id
stringlengths
9
10
submitter
stringlengths
1
64
authors
stringlengths
4
20.7k
title
stringlengths
4
246
comments
stringlengths
1
523
journal-ref
stringlengths
4
404
doi
stringlengths
11
153
report-no
stringlengths
2
254
categories
stringlengths
5
98
license
stringclasses
9 values
orig_abstract
stringlengths
14
3.35k
versions
listlengths
1
60
update_date
stringlengths
10
10
authors_parsed
listlengths
1
1.35k
abstract
stringlengths
11
3.34k
2107.01799
Isaac Sledge
Isaac J. Sledge and Jose C. Principe
An Information-Theoretic Approach for Automatically Determining the Number of States when Aggregating Markov Chains
Submitted to IEEE ICASSP. arXiv admin note: substantial text overlap with arXiv:1903.09266
null
10.1109/ICASSP.2019.8682473
null
cs.IT cs.LG math.IT
http://creativecommons.org/licenses/by/4.0/
A fundamental problem when aggregating Markov chains is the specification of the number of state groups. Too few state groups may fail to sufficiently capture the pertinent dynamics of the original, high-order Markov chain. Too many state groups may lead to a non-parsimonious, reduced-order Markov chain whose complexity rivals that of the original. In this paper, we show that an augmented value-of-information-based approach to aggregating Markov chains facilitates the determination of the number of state groups. The optimal state-group count coincides with the case where the complexity of the reduced-order chain is balanced against the mutual dependence between the original- and reduced-order chain dynamics.
[ { "created": "Mon, 5 Jul 2021 05:36:04 GMT", "version": "v1" } ]
2021-07-06
[ [ "Sledge", "Isaac J.", "" ], [ "Principe", "Jose C.", "" ] ]
A fundamental problem when aggregating Markov chains is the specification of the number of state groups. Too few state groups may fail to sufficiently capture the pertinent dynamics of the original, high-order Markov chain. Too many state groups may lead to a non-parsimonious, reduced-order Markov chain whose complexity rivals that of the original. In this paper, we show that an augmented value-of-information-based approach to aggregating Markov chains facilitates the determination of the number of state groups. The optimal state-group count coincides with the case where the complexity of the reduced-order chain is balanced against the mutual dependence between the original- and reduced-order chain dynamics.
2102.01884
Bekir Sait Ciftler
Bekir Sait Ciftler, Abdulmalik Alwarafy, Mohamed Abdallah, Mounir Hamdi
DQN-Based Multi-User Power Allocation for Hybrid RF/VLC Networks
6 pages, 4 figures, accepted to IEEE ICC 2021
null
null
null
cs.NI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper, a Deep Q-Network (DQN) based multi-agent multi-user power allocation algorithm is proposed for hybrid networks composed of radio frequency (RF) and visible light communication (VLC) access points (APs). The users are capable of multihoming, which can bridge RF and VLC links for accommodating their bandwidth requirements. By leveraging a non-cooperative multi-agent DQN algorithm, where each AP is an agent, an online power allocation strategy is developed to optimize the transmit power for providing users' required data rate. Our simulation results demonstrate that DQN's median convergence time training is 90% shorter than the Q-Learning (QL) based algorithm. The DQN-based algorithm converges to the desired user rate in half duration on average while converging with the rate of 96.1% compared to the QL-based algorithm's convergence rate of 72.3% Additionally, thanks to its continuous state-space definition, the DQN-based power allocation algorithm provides average user data rates closer to the target rates than the QL-based algorithm when it converges.
[ { "created": "Wed, 3 Feb 2021 05:42:49 GMT", "version": "v1" } ]
2021-02-04
[ [ "Ciftler", "Bekir Sait", "" ], [ "Alwarafy", "Abdulmalik", "" ], [ "Abdallah", "Mohamed", "" ], [ "Hamdi", "Mounir", "" ] ]
In this paper, a Deep Q-Network (DQN) based multi-agent multi-user power allocation algorithm is proposed for hybrid networks composed of radio frequency (RF) and visible light communication (VLC) access points (APs). The users are capable of multihoming, which can bridge RF and VLC links for accommodating their bandwidth requirements. By leveraging a non-cooperative multi-agent DQN algorithm, where each AP is an agent, an online power allocation strategy is developed to optimize the transmit power for providing users' required data rate. Our simulation results demonstrate that DQN's median convergence time training is 90% shorter than the Q-Learning (QL) based algorithm. The DQN-based algorithm converges to the desired user rate in half duration on average while converging with the rate of 96.1% compared to the QL-based algorithm's convergence rate of 72.3% Additionally, thanks to its continuous state-space definition, the DQN-based power allocation algorithm provides average user data rates closer to the target rates than the QL-based algorithm when it converges.
1904.07331
Niki Gitinabard
Adithya Sheshadri, Niki Gitinabard, Collin F. Lynch, Tiffany Barnes, and Sarah Heckman
Predicting Student Performance Based on Online Study Habits: A Study of Blended Courses
Published in the International Conference on Educational Data Mining (EDM 2018)
null
null
null
cs.CY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Online tools provide unique access to research students' study habits and problem-solving behavior. In MOOCs, this online data can be used to inform instructors and to provide automatic guidance to students. However, these techniques may not apply in blended courses with face to face and online components. We report on a study of integrated user-system interaction logs from 3 computer science courses using four online systems: LMS, forum, version control, and homework system. Our results show that students rarely work across platforms in a single session, and that final class performance can be predicted from students' system use.
[ { "created": "Mon, 15 Apr 2019 21:18:13 GMT", "version": "v1" } ]
2019-04-17
[ [ "Sheshadri", "Adithya", "" ], [ "Gitinabard", "Niki", "" ], [ "Lynch", "Collin F.", "" ], [ "Barnes", "Tiffany", "" ], [ "Heckman", "Sarah", "" ] ]
Online tools provide unique access to research students' study habits and problem-solving behavior. In MOOCs, this online data can be used to inform instructors and to provide automatic guidance to students. However, these techniques may not apply in blended courses with face to face and online components. We report on a study of integrated user-system interaction logs from 3 computer science courses using four online systems: LMS, forum, version control, and homework system. Our results show that students rarely work across platforms in a single session, and that final class performance can be predicted from students' system use.
2107.06552
Young Eun Kim
Young Eun Kim and Seong-Whan Lee
Domain Generalization with Pseudo-Domain Label for Face Anti-Spoofing
null
null
null
null
cs.CV cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Face anti-spoofing (FAS) plays an important role in protecting face recognition systems from face representation attacks. Many recent studies in FAS have approached this problem with domain generalization technique. Domain generalization aims to increase generalization performance to better detect various types of attacks and unseen attacks. However, previous studies in this area have defined each domain simply as an anti-spoofing datasets and focused on developing learning techniques. In this paper, we proposed a method that enables network to judge its domain by itself with the clustered convolutional feature statistics from intermediate layers of the network, without labeling domains as datasets. We obtained pseudo-domain labels by not only using the network extracting features, but also using depth estimators, which were previously used only as an auxiliary task in FAS. In our experiments, we trained with three datasets and evaluated the performance with the remaining one dataset to demonstrate the effectiveness of the proposed method by conducting a total of four sets of experiments.
[ { "created": "Wed, 14 Jul 2021 08:35:07 GMT", "version": "v1" } ]
2021-07-15
[ [ "Kim", "Young Eun", "" ], [ "Lee", "Seong-Whan", "" ] ]
Face anti-spoofing (FAS) plays an important role in protecting face recognition systems from face representation attacks. Many recent studies in FAS have approached this problem with domain generalization technique. Domain generalization aims to increase generalization performance to better detect various types of attacks and unseen attacks. However, previous studies in this area have defined each domain simply as an anti-spoofing datasets and focused on developing learning techniques. In this paper, we proposed a method that enables network to judge its domain by itself with the clustered convolutional feature statistics from intermediate layers of the network, without labeling domains as datasets. We obtained pseudo-domain labels by not only using the network extracting features, but also using depth estimators, which were previously used only as an auxiliary task in FAS. In our experiments, we trained with three datasets and evaluated the performance with the remaining one dataset to demonstrate the effectiveness of the proposed method by conducting a total of four sets of experiments.
2004.13297
Jian Ren
Menglei Chai, Jian Ren, Sergey Tulyakov
Neural Hair Rendering
ECCV 2020
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper, we propose a generic neural-based hair rendering pipeline that can synthesize photo-realistic images from virtual 3D hair models. Unlike existing supervised translation methods that require model-level similarity to preserve consistent structure representation for both real images and fake renderings, our method adopts an unsupervised solution to work on arbitrary hair models. The key component of our method is a shared latent space to encode appearance-invariant structure information of both domains, which generates realistic renderings conditioned by extra appearance inputs. This is achieved by domain-specific pre-disentangled structure representation, partially shared domain encoder layers and a structure discriminator. We also propose a simple yet effective temporal conditioning method to enforce consistency for video sequence generation. We demonstrate the superiority of our method by testing it on a large number of portraits and comparing it with alternative baselines and state-of-the-art unsupervised image translation methods.
[ { "created": "Tue, 28 Apr 2020 04:36:49 GMT", "version": "v1" }, { "created": "Tue, 21 Jul 2020 19:29:30 GMT", "version": "v2" } ]
2020-07-23
[ [ "Chai", "Menglei", "" ], [ "Ren", "Jian", "" ], [ "Tulyakov", "Sergey", "" ] ]
In this paper, we propose a generic neural-based hair rendering pipeline that can synthesize photo-realistic images from virtual 3D hair models. Unlike existing supervised translation methods that require model-level similarity to preserve consistent structure representation for both real images and fake renderings, our method adopts an unsupervised solution to work on arbitrary hair models. The key component of our method is a shared latent space to encode appearance-invariant structure information of both domains, which generates realistic renderings conditioned by extra appearance inputs. This is achieved by domain-specific pre-disentangled structure representation, partially shared domain encoder layers and a structure discriminator. We also propose a simple yet effective temporal conditioning method to enforce consistency for video sequence generation. We demonstrate the superiority of our method by testing it on a large number of portraits and comparing it with alternative baselines and state-of-the-art unsupervised image translation methods.
1811.12640
Sudeshna Roy
Sudeshna Roy, Meghana Madhyastha, Sheril Lawrence, Vaibhav Rajan
Inferring Concept Prerequisite Relations from Online Educational Resources
Accepted at the AAAI Conference on Innovative Applications of Artificial Intelligence (IAAI-19)
null
null
null
cs.CL cs.AI cs.LG stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The Internet has rich and rapidly increasing sources of high quality educational content. Inferring prerequisite relations between educational concepts is required for modern large-scale online educational technology applications such as personalized recommendations and automatic curriculum creation. We present PREREQ, a new supervised learning method for inferring concept prerequisite relations. PREREQ is designed using latent representations of concepts obtained from the Pairwise Latent Dirichlet Allocation model, and a neural network based on the Siamese network architecture. PREREQ can learn unknown concept prerequisites from course prerequisites and labeled concept prerequisite data. It outperforms state-of-the-art approaches on benchmark datasets and can effectively learn from very less training data. PREREQ can also use unlabeled video playlists, a steadily growing source of training data, to learn concept prerequisites, thus obviating the need for manual annotation of course prerequisites.
[ { "created": "Fri, 30 Nov 2018 06:55:20 GMT", "version": "v1" }, { "created": "Wed, 23 Jan 2019 00:39:00 GMT", "version": "v2" } ]
2019-01-24
[ [ "Roy", "Sudeshna", "" ], [ "Madhyastha", "Meghana", "" ], [ "Lawrence", "Sheril", "" ], [ "Rajan", "Vaibhav", "" ] ]
The Internet has rich and rapidly increasing sources of high quality educational content. Inferring prerequisite relations between educational concepts is required for modern large-scale online educational technology applications such as personalized recommendations and automatic curriculum creation. We present PREREQ, a new supervised learning method for inferring concept prerequisite relations. PREREQ is designed using latent representations of concepts obtained from the Pairwise Latent Dirichlet Allocation model, and a neural network based on the Siamese network architecture. PREREQ can learn unknown concept prerequisites from course prerequisites and labeled concept prerequisite data. It outperforms state-of-the-art approaches on benchmark datasets and can effectively learn from very less training data. PREREQ can also use unlabeled video playlists, a steadily growing source of training data, to learn concept prerequisites, thus obviating the need for manual annotation of course prerequisites.
1910.14634
Reinhard Heckel
Reinhard Heckel and Mahdi Soltanolkotabi
Denoising and Regularization via Exploiting the Structural Bias of Convolutional Generators
final ICRL version; simplifications in the proof
International Conference on Learning Representations (ICLR) 2020
null
null
cs.LG cs.CV cs.IT math.IT stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Convolutional Neural Networks (CNNs) have emerged as highly successful tools for image generation, recovery, and restoration. A major contributing factor to this success is that convolutional networks impose strong prior assumptions about natural images. A surprising experiment that highlights this architectural bias towards natural images is that one can remove noise and corruptions from a natural image without using any training data, by simply fitting (via gradient descent) a randomly initialized, over-parameterized convolutional generator to the corrupted image. While this over-parameterized network can fit the corrupted image perfectly, surprisingly after a few iterations of gradient descent it generates an almost uncorrupted image. This intriguing phenomenon enables state-of-the-art CNN-based denoising and regularization of other inverse problems. In this paper, we attribute this effect to a particular architectural choice of convolutional networks, namely convolutions with fixed interpolating filters. We then formally characterize the dynamics of fitting a two-layer convolutional generator to a noisy signal and prove that early-stopped gradient descent denoises/regularizes. Our proof relies on showing that convolutional generators fit the structured part of an image significantly faster than the corrupted portion.
[ { "created": "Thu, 31 Oct 2019 17:22:00 GMT", "version": "v1" }, { "created": "Sun, 23 Feb 2020 01:49:25 GMT", "version": "v2" } ]
2020-02-25
[ [ "Heckel", "Reinhard", "" ], [ "Soltanolkotabi", "Mahdi", "" ] ]
Convolutional Neural Networks (CNNs) have emerged as highly successful tools for image generation, recovery, and restoration. A major contributing factor to this success is that convolutional networks impose strong prior assumptions about natural images. A surprising experiment that highlights this architectural bias towards natural images is that one can remove noise and corruptions from a natural image without using any training data, by simply fitting (via gradient descent) a randomly initialized, over-parameterized convolutional generator to the corrupted image. While this over-parameterized network can fit the corrupted image perfectly, surprisingly after a few iterations of gradient descent it generates an almost uncorrupted image. This intriguing phenomenon enables state-of-the-art CNN-based denoising and regularization of other inverse problems. In this paper, we attribute this effect to a particular architectural choice of convolutional networks, namely convolutions with fixed interpolating filters. We then formally characterize the dynamics of fitting a two-layer convolutional generator to a noisy signal and prove that early-stopped gradient descent denoises/regularizes. Our proof relies on showing that convolutional generators fit the structured part of an image significantly faster than the corrupted portion.
2307.03638
Jianyuan Ni
Jianyuan Ni, Hao Tang, Anne H.H. Ngu, Gaowen Liu, Yan Yan
Physical-aware Cross-modal Adversarial Network for Wearable Sensor-based Human Action Recognition
We will be making some significant changes to the paper, including the title and methodology. We therefore wish to withdraw the paper for now
null
null
null
cs.MM cs.HC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Wearable sensor-based Human Action Recognition (HAR) has made significant strides in recent times. However, the accuracy performance of wearable sensor-based HAR is currently still lagging behind that of visual modalities-based systems, such as RGB video and depth data. Although diverse input modalities can provide complementary cues and improve the accuracy performance of HAR, wearable devices can only capture limited kinds of non-visual time series input, such as accelerometers and gyroscopes. This limitation hinders the deployment of multimodal simultaneously using visual and non-visual modality data in parallel on current wearable devices. To address this issue, we propose a novel Physical-aware Cross-modal Adversarial (PCA) framework that utilizes only time-series accelerometer data from four inertial sensors for the wearable sensor-based HAR problem. Specifically, we propose an effective IMU2SKELETON network to produce corresponding synthetic skeleton joints from accelerometer data. Subsequently, we imposed additional constraints on the synthetic skeleton data from a physical perspective, as accelerometer data can be regarded as the second derivative of the skeleton sequence coordinates. After that, the original accelerometer as well as the constrained skeleton sequence were fused together to make the final classification. In this way, when individuals wear wearable devices, the devices can not only capture accelerometer data, but can also generate synthetic skeleton sequences for real-time wearable sensor-based HAR applications that need to be conducted anytime and anywhere. To demonstrate the effectiveness of our proposed PCA framework, we conduct extensive experiments on Berkeley-MHAD, UTD-MHAD, and MMAct datasets. The results confirm that the proposed PCA approach has competitive performance compared to the previous methods on the mono sensor-based HAR classification problem.
[ { "created": "Fri, 7 Jul 2023 14:57:34 GMT", "version": "v1" }, { "created": "Sun, 19 May 2024 20:39:25 GMT", "version": "v2" } ]
2024-05-21
[ [ "Ni", "Jianyuan", "" ], [ "Tang", "Hao", "" ], [ "Ngu", "Anne H. H.", "" ], [ "Liu", "Gaowen", "" ], [ "Yan", "Yan", "" ] ]
Wearable sensor-based Human Action Recognition (HAR) has made significant strides in recent times. However, the accuracy performance of wearable sensor-based HAR is currently still lagging behind that of visual modalities-based systems, such as RGB video and depth data. Although diverse input modalities can provide complementary cues and improve the accuracy performance of HAR, wearable devices can only capture limited kinds of non-visual time series input, such as accelerometers and gyroscopes. This limitation hinders the deployment of multimodal simultaneously using visual and non-visual modality data in parallel on current wearable devices. To address this issue, we propose a novel Physical-aware Cross-modal Adversarial (PCA) framework that utilizes only time-series accelerometer data from four inertial sensors for the wearable sensor-based HAR problem. Specifically, we propose an effective IMU2SKELETON network to produce corresponding synthetic skeleton joints from accelerometer data. Subsequently, we imposed additional constraints on the synthetic skeleton data from a physical perspective, as accelerometer data can be regarded as the second derivative of the skeleton sequence coordinates. After that, the original accelerometer as well as the constrained skeleton sequence were fused together to make the final classification. In this way, when individuals wear wearable devices, the devices can not only capture accelerometer data, but can also generate synthetic skeleton sequences for real-time wearable sensor-based HAR applications that need to be conducted anytime and anywhere. To demonstrate the effectiveness of our proposed PCA framework, we conduct extensive experiments on Berkeley-MHAD, UTD-MHAD, and MMAct datasets. The results confirm that the proposed PCA approach has competitive performance compared to the previous methods on the mono sensor-based HAR classification problem.
1908.02802
Roozbeh Yousefzadeh
Roozbeh Yousefzadeh, Dianne P O'Leary
Investigating Decision Boundaries of Trained Neural Networks
null
null
null
null
cs.LG stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Deep learning models have been the subject of study from various perspectives, for example, their training process, interpretation, generalization error, robustness to adversarial attacks, etc. A trained model is defined by its decision boundaries, and therefore, many of the studies about deep learning models speculate about the decision boundaries, and sometimes make simplifying assumptions about them. So far, finding exact points on the decision boundaries of trained deep models has been considered an intractable problem. Here, we compute exact points on the decision boundaries of these models and provide mathematical tools to investigate the surfaces that define the decision boundaries. Through numerical results, we confirm that some of the speculations about the decision boundaries are accurate, some of the computational methods can be improved, and some of the simplifying assumptions may be unreliable, for models with nonlinear activation functions. We advocate for verification of simplifying assumptions and approximation methods, wherever they are used. Finally, we demonstrate that the computational practices used for finding adversarial examples can be improved and computing the closest point on the decision boundary reveals the weakest vulnerability of a model against adversarial attack.
[ { "created": "Wed, 7 Aug 2019 19:09:22 GMT", "version": "v1" } ]
2019-08-09
[ [ "Yousefzadeh", "Roozbeh", "" ], [ "O'Leary", "Dianne P", "" ] ]
Deep learning models have been the subject of study from various perspectives, for example, their training process, interpretation, generalization error, robustness to adversarial attacks, etc. A trained model is defined by its decision boundaries, and therefore, many of the studies about deep learning models speculate about the decision boundaries, and sometimes make simplifying assumptions about them. So far, finding exact points on the decision boundaries of trained deep models has been considered an intractable problem. Here, we compute exact points on the decision boundaries of these models and provide mathematical tools to investigate the surfaces that define the decision boundaries. Through numerical results, we confirm that some of the speculations about the decision boundaries are accurate, some of the computational methods can be improved, and some of the simplifying assumptions may be unreliable, for models with nonlinear activation functions. We advocate for verification of simplifying assumptions and approximation methods, wherever they are used. Finally, we demonstrate that the computational practices used for finding adversarial examples can be improved and computing the closest point on the decision boundary reveals the weakest vulnerability of a model against adversarial attack.
2111.08211
Yan Kang
Yuezhou Wu, Yan Kang, Jiahuan Luo, Yuanqin He, Qiang Yang
FedCG: Leverage Conditional GAN for Protecting Privacy and Maintaining Competitive Performance in Federated Learning
null
Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, 2022
10.24963/ijcai.2022/324
null
cs.LG
http://creativecommons.org/licenses/by/4.0/
Federated learning (FL) aims to protect data privacy by enabling clients to build machine learning models collaboratively without sharing their private data. Recent works demonstrate that information exchanged during FL is subject to gradient-based privacy attacks, and consequently, a variety of privacy-preserving methods have been adopted to thwart such attacks. However, these defensive methods either introduce orders of magnitude more computational and communication overheads (e.g., with homomorphic encryption) or incur substantial model performance losses in terms of prediction accuracy (e.g., with differential privacy). In this work, we propose $\textsc{FedCG}$, a novel federated learning method that leverages conditional generative adversarial networks to achieve high-level privacy protection while still maintaining competitive model performance. $\textsc{FedCG}$ decomposes each client's local network into a private extractor and a public classifier and keeps the extractor local to protect privacy. Instead of exposing extractors, $\textsc{FedCG}$ shares clients' generators with the server for aggregating clients' shared knowledge, aiming to enhance the performance of each client's local networks. Extensive experiments demonstrate that $\textsc{FedCG}$ can achieve competitive model performance compared with FL baselines, and privacy analysis shows that $\textsc{FedCG}$ has a high-level privacy-preserving capability. Code is available at https://github.com/yankang18/FedCG
[ { "created": "Tue, 16 Nov 2021 03:20:37 GMT", "version": "v1" }, { "created": "Wed, 16 Feb 2022 03:16:52 GMT", "version": "v2" }, { "created": "Sun, 7 Jul 2024 03:57:12 GMT", "version": "v3" } ]
2024-07-09
[ [ "Wu", "Yuezhou", "" ], [ "Kang", "Yan", "" ], [ "Luo", "Jiahuan", "" ], [ "He", "Yuanqin", "" ], [ "Yang", "Qiang", "" ] ]
Federated learning (FL) aims to protect data privacy by enabling clients to build machine learning models collaboratively without sharing their private data. Recent works demonstrate that information exchanged during FL is subject to gradient-based privacy attacks, and consequently, a variety of privacy-preserving methods have been adopted to thwart such attacks. However, these defensive methods either introduce orders of magnitude more computational and communication overheads (e.g., with homomorphic encryption) or incur substantial model performance losses in terms of prediction accuracy (e.g., with differential privacy). In this work, we propose $\textsc{FedCG}$, a novel federated learning method that leverages conditional generative adversarial networks to achieve high-level privacy protection while still maintaining competitive model performance. $\textsc{FedCG}$ decomposes each client's local network into a private extractor and a public classifier and keeps the extractor local to protect privacy. Instead of exposing extractors, $\textsc{FedCG}$ shares clients' generators with the server for aggregating clients' shared knowledge, aiming to enhance the performance of each client's local networks. Extensive experiments demonstrate that $\textsc{FedCG}$ can achieve competitive model performance compared with FL baselines, and privacy analysis shows that $\textsc{FedCG}$ has a high-level privacy-preserving capability. Code is available at https://github.com/yankang18/FedCG
1208.0079
Abhay Jha
Abhay Jha, Dan Suciu
Probabilistic Databases with MarkoViews
VLDB2012
Proceedings of the VLDB Endowment (PVLDB), Vol. 5, No. 11, pp. 1160-1171 (2012)
null
null
cs.DB
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Most of the work on query evaluation in probabilistic databases has focused on the simple tuple-independent data model, where tuples are independent random events. Several efficient query evaluation techniques exists in this setting, such as safe plans, algorithms based on OBDDs, tree-decomposition and a variety of approximation algorithms. However, complex data analytics tasks often require complex correlations, and query evaluation then is significantly more expensive, or more restrictive. In this paper, we propose MVDB as a framework both for representing complex correlations and for efficient query evaluation. An MVDB specifies correlations by views, called MarkoViews, on the probabilistic relations and declaring the weights of the view's outputs. An MVDB is a (very large) Markov Logic Network. We make two sets of contributions. First, we show that query evaluation on an MVDB is equivalent to evaluating a Union of Conjunctive Query(UCQ) over a tuple-independent database. The translation is exact (thus allowing the techniques developed for tuple independent databases to be carried over to MVDB), yet it is novel and quite non-obvious (some resulting probabilities may be negative!). This translation in itself though may not lead to much gain since the translated query gets complicated as we try to capture more correlations. Our second contribution is to propose a new query evaluation strategy that exploits offline compilation to speed up online query evaluation. Here we utilize and extend our prior work on compilation of UCQ. We validate experimentally our techniques on a large probabilistic database with MarkoViews inferred from the DBLP data.
[ { "created": "Wed, 1 Aug 2012 03:47:10 GMT", "version": "v1" } ]
2012-08-02
[ [ "Jha", "Abhay", "" ], [ "Suciu", "Dan", "" ] ]
Most of the work on query evaluation in probabilistic databases has focused on the simple tuple-independent data model, where tuples are independent random events. Several efficient query evaluation techniques exists in this setting, such as safe plans, algorithms based on OBDDs, tree-decomposition and a variety of approximation algorithms. However, complex data analytics tasks often require complex correlations, and query evaluation then is significantly more expensive, or more restrictive. In this paper, we propose MVDB as a framework both for representing complex correlations and for efficient query evaluation. An MVDB specifies correlations by views, called MarkoViews, on the probabilistic relations and declaring the weights of the view's outputs. An MVDB is a (very large) Markov Logic Network. We make two sets of contributions. First, we show that query evaluation on an MVDB is equivalent to evaluating a Union of Conjunctive Query(UCQ) over a tuple-independent database. The translation is exact (thus allowing the techniques developed for tuple independent databases to be carried over to MVDB), yet it is novel and quite non-obvious (some resulting probabilities may be negative!). This translation in itself though may not lead to much gain since the translated query gets complicated as we try to capture more correlations. Our second contribution is to propose a new query evaluation strategy that exploits offline compilation to speed up online query evaluation. Here we utilize and extend our prior work on compilation of UCQ. We validate experimentally our techniques on a large probabilistic database with MarkoViews inferred from the DBLP data.
1902.09093
Bishan Yang
Igor Labutov, Bishan Yang, Anusha Prakash, Amos Azaria
Multi-Relational Question Answering from Narratives: Machine Reading and Reasoning in Simulated Worlds
published at ACL 2018
ACL 2018
null
null
cs.CL cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Question Answering (QA), as a research field, has primarily focused on either knowledge bases (KBs) or free text as a source of knowledge. These two sources have historically shaped the kinds of questions that are asked over these sources, and the methods developed to answer them. In this work, we look towards a practical use-case of QA over user-instructed knowledge that uniquely combines elements of both structured QA over knowledge bases, and unstructured QA over narrative, introducing the task of multi-relational QA over personal narrative. As a first step towards this goal, we make three key contributions: (i) we generate and release TextWorldsQA, a set of five diverse datasets, where each dataset contains dynamic narrative that describes entities and relations in a simulated world, paired with variably compositional questions over that knowledge, (ii) we perform a thorough evaluation and analysis of several state-of-the-art QA models and their variants at this task, and (iii) we release a lightweight Python-based framework we call TextWorlds for easily generating arbitrary additional worlds and narrative, with the goal of allowing the community to create and share a growing collection of diverse worlds as a test-bed for this task.
[ { "created": "Mon, 25 Feb 2019 05:04:26 GMT", "version": "v1" } ]
2019-02-26
[ [ "Labutov", "Igor", "" ], [ "Yang", "Bishan", "" ], [ "Prakash", "Anusha", "" ], [ "Azaria", "Amos", "" ] ]
Question Answering (QA), as a research field, has primarily focused on either knowledge bases (KBs) or free text as a source of knowledge. These two sources have historically shaped the kinds of questions that are asked over these sources, and the methods developed to answer them. In this work, we look towards a practical use-case of QA over user-instructed knowledge that uniquely combines elements of both structured QA over knowledge bases, and unstructured QA over narrative, introducing the task of multi-relational QA over personal narrative. As a first step towards this goal, we make three key contributions: (i) we generate and release TextWorldsQA, a set of five diverse datasets, where each dataset contains dynamic narrative that describes entities and relations in a simulated world, paired with variably compositional questions over that knowledge, (ii) we perform a thorough evaluation and analysis of several state-of-the-art QA models and their variants at this task, and (iii) we release a lightweight Python-based framework we call TextWorlds for easily generating arbitrary additional worlds and narrative, with the goal of allowing the community to create and share a growing collection of diverse worlds as a test-bed for this task.
1710.02555
Melissa Greeff
Melissa Greeff and Angela P. Schoellig
Model Predictive Path-Following for Constrained Differentially Flat Systems
8 pages, submitted to ICRA 2018
null
null
null
cs.RO cs.SY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
For many tasks, predictive path-following control can significantly improve the performance and robustness of autonomous robots over traditional trajectory tracking control. It does this by prioritizing closeness to the path over timed progress along the path and by looking ahead to account for changes in the path. We propose a novel predictive path-following approach that couples feedforward linearization with path-based model predictive control. Our approach has a few key advantages. By utilizing the differential flatness property, we reduce the path-based model predictive control problem from a nonlinear to a convex optimization problem. Robustness to disturbances is achieved by a dynamic path reference, which adjusts its speed based on the robot's progress. We also account for key system constraints. We demonstrate these advantages in experiment on a quadrotor. We show improved performance over a baseline trajectory tracking controller by keeping the quadrotor closer to the desired path under nominal conditions, with an initial offset and under a wind disturbance.
[ { "created": "Fri, 6 Oct 2017 18:56:13 GMT", "version": "v1" }, { "created": "Thu, 2 Nov 2017 15:24:08 GMT", "version": "v2" } ]
2017-11-03
[ [ "Greeff", "Melissa", "" ], [ "Schoellig", "Angela P.", "" ] ]
For many tasks, predictive path-following control can significantly improve the performance and robustness of autonomous robots over traditional trajectory tracking control. It does this by prioritizing closeness to the path over timed progress along the path and by looking ahead to account for changes in the path. We propose a novel predictive path-following approach that couples feedforward linearization with path-based model predictive control. Our approach has a few key advantages. By utilizing the differential flatness property, we reduce the path-based model predictive control problem from a nonlinear to a convex optimization problem. Robustness to disturbances is achieved by a dynamic path reference, which adjusts its speed based on the robot's progress. We also account for key system constraints. We demonstrate these advantages in experiment on a quadrotor. We show improved performance over a baseline trajectory tracking controller by keeping the quadrotor closer to the desired path under nominal conditions, with an initial offset and under a wind disturbance.
2008.09511
Peter Lindner
Nofar Carmeli, Martin Grohe, Peter Lindner, Christoph Standke
Tuple-Independent Representations of Infinite Probabilistic Databases
null
null
null
null
cs.DB cs.LO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Probabilistic databases (PDBs) are probability spaces over database instances. They provide a framework for handling uncertainty in databases, as occurs due to data integration, noisy data, data from unreliable sources or randomized processes. Most of the existing theory literature investigated finite, tuple-independent PDBs (TI-PDBs) where the occurrences of tuples are independent events. Only recently, Grohe and Lindner (PODS '19) introduced independence assumptions for PDBs beyond the finite domain assumption. In the finite, a major argument for discussing the theoretical properties of TI-PDBs is that they can be used to represent any finite PDB via views. This is no longer the case once the number of tuples is countably infinite. In this paper, we systematically study the representability of infinite PDBs in terms of TI-PDBs and the related block-independent disjoint PDBs. The central question is which infinite PDBs are representable as first-order views over tuple-independent PDBs. We give a necessary condition for the representability of PDBs and provide a sufficient criterion for representability in terms of the probability distribution of a PDB. With various examples, we explore the limits of our criteria. We show that conditioning on first order properties yields no additional power in terms of expressivity. Finally, we discuss the relation between purely logical and arithmetic reasons for (non-)representability.
[ { "created": "Fri, 21 Aug 2020 14:39:47 GMT", "version": "v1" }, { "created": "Tue, 19 Apr 2022 07:17:45 GMT", "version": "v2" } ]
2022-04-20
[ [ "Carmeli", "Nofar", "" ], [ "Grohe", "Martin", "" ], [ "Lindner", "Peter", "" ], [ "Standke", "Christoph", "" ] ]
Probabilistic databases (PDBs) are probability spaces over database instances. They provide a framework for handling uncertainty in databases, as occurs due to data integration, noisy data, data from unreliable sources or randomized processes. Most of the existing theory literature investigated finite, tuple-independent PDBs (TI-PDBs) where the occurrences of tuples are independent events. Only recently, Grohe and Lindner (PODS '19) introduced independence assumptions for PDBs beyond the finite domain assumption. In the finite, a major argument for discussing the theoretical properties of TI-PDBs is that they can be used to represent any finite PDB via views. This is no longer the case once the number of tuples is countably infinite. In this paper, we systematically study the representability of infinite PDBs in terms of TI-PDBs and the related block-independent disjoint PDBs. The central question is which infinite PDBs are representable as first-order views over tuple-independent PDBs. We give a necessary condition for the representability of PDBs and provide a sufficient criterion for representability in terms of the probability distribution of a PDB. With various examples, we explore the limits of our criteria. We show that conditioning on first order properties yields no additional power in terms of expressivity. Finally, we discuss the relation between purely logical and arithmetic reasons for (non-)representability.
2302.08174
Rafael Mohr
Christian Eder, Pierre Lairez, Rafael Mohr, Mohab Safey El Din
A Direttissimo Algorithm for Equidimensional Decomposition
Some minor revisions, corrects a mistake in the proof of lemma 2.2
null
null
null
cs.SC math.AC
http://creativecommons.org/licenses/by/4.0/
We describe a recursive algorithm that decomposes an algebraic set into locally closed equidimensional sets, i.e. sets which each have irreducible components of the same dimension. At the core of this algorithm, we combine ideas from the theory of triangular sets, a.k.a. regular chains, with Gr\"obner bases to encode and work with locally closed algebraic sets. Equipped with this, our algorithm avoids projections of the algebraic sets that are decomposed and certain genericity assumptions frequently made when decomposing polynomial systems, such as assumptions about Noether position. This makes it produce fine decompositions on more structured systems where ensuring genericity assumptions often destroys the structure of the system at hand. Practical experiments demonstrate its efficiency compared to state-of-the-art implementations.
[ { "created": "Thu, 16 Feb 2023 09:42:55 GMT", "version": "v1" }, { "created": "Fri, 9 Jun 2023 08:45:15 GMT", "version": "v2" } ]
2023-06-12
[ [ "Eder", "Christian", "" ], [ "Lairez", "Pierre", "" ], [ "Mohr", "Rafael", "" ], [ "Din", "Mohab Safey El", "" ] ]
We describe a recursive algorithm that decomposes an algebraic set into locally closed equidimensional sets, i.e. sets which each have irreducible components of the same dimension. At the core of this algorithm, we combine ideas from the theory of triangular sets, a.k.a. regular chains, with Gr\"obner bases to encode and work with locally closed algebraic sets. Equipped with this, our algorithm avoids projections of the algebraic sets that are decomposed and certain genericity assumptions frequently made when decomposing polynomial systems, such as assumptions about Noether position. This makes it produce fine decompositions on more structured systems where ensuring genericity assumptions often destroys the structure of the system at hand. Practical experiments demonstrate its efficiency compared to state-of-the-art implementations.
2406.12479
Haifeng Li
Linrui Xu, Ling Zhao, Wang Guo, Qiujun Li, Kewang Long, Kaiqi Zou, Yuhan Wang, Haifeng Li
RS-GPT4V: A Unified Multimodal Instruction-Following Dataset for Remote Sensing Image Understanding
14 pages, 6 figures, 4 tables
null
null
null
cs.CV cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The remote sensing image intelligence understanding model is undergoing a new profound paradigm shift which has been promoted by multi-modal large language model (MLLM), i.e. from the paradigm learning a domain model (LaDM) shifts to paradigm learning a pre-trained general foundation model followed by an adaptive domain model (LaGD). Under the new LaGD paradigm, the old datasets, which have led to advances in RSI intelligence understanding in the last decade, are no longer suitable for fire-new tasks. We argued that a new dataset must be designed to lighten tasks with the following features: 1) Generalization: training model to learn shared knowledge among tasks and to adapt to different tasks; 2) Understanding complex scenes: training model to understand the fine-grained attribute of the objects of interest, and to be able to describe the scene with natural language; 3) Reasoning: training model to be able to realize high-level visual reasoning. In this paper, we designed a high-quality, diversified, and unified multimodal instruction-following dataset for RSI understanding produced by GPT-4V and existing datasets, which we called RS-GPT4V. To achieve generalization, we used a (Question, Answer) which was deduced from GPT-4V via instruction-following to unify the tasks such as captioning and localization; To achieve complex scene, we proposed a hierarchical instruction description with local strategy in which the fine-grained attributes of the objects and their spatial relationships are described and global strategy in which all the local information are integrated to yield detailed instruction descript; To achieve reasoning, we designed multiple-turn QA pair to provide the reasoning ability for a model. The empirical results show that the fine-tuned MLLMs by RS-GPT4V can describe fine-grained information. The dataset is available at: https://github.com/GeoX-Lab/RS-GPT4V.
[ { "created": "Tue, 18 Jun 2024 10:34:28 GMT", "version": "v1" } ]
2024-06-19
[ [ "Xu", "Linrui", "" ], [ "Zhao", "Ling", "" ], [ "Guo", "Wang", "" ], [ "Li", "Qiujun", "" ], [ "Long", "Kewang", "" ], [ "Zou", "Kaiqi", "" ], [ "Wang", "Yuhan", "" ], [ "Li", "Haifeng", "" ] ]
The remote sensing image intelligence understanding model is undergoing a new profound paradigm shift which has been promoted by multi-modal large language model (MLLM), i.e. from the paradigm learning a domain model (LaDM) shifts to paradigm learning a pre-trained general foundation model followed by an adaptive domain model (LaGD). Under the new LaGD paradigm, the old datasets, which have led to advances in RSI intelligence understanding in the last decade, are no longer suitable for fire-new tasks. We argued that a new dataset must be designed to lighten tasks with the following features: 1) Generalization: training model to learn shared knowledge among tasks and to adapt to different tasks; 2) Understanding complex scenes: training model to understand the fine-grained attribute of the objects of interest, and to be able to describe the scene with natural language; 3) Reasoning: training model to be able to realize high-level visual reasoning. In this paper, we designed a high-quality, diversified, and unified multimodal instruction-following dataset for RSI understanding produced by GPT-4V and existing datasets, which we called RS-GPT4V. To achieve generalization, we used a (Question, Answer) which was deduced from GPT-4V via instruction-following to unify the tasks such as captioning and localization; To achieve complex scene, we proposed a hierarchical instruction description with local strategy in which the fine-grained attributes of the objects and their spatial relationships are described and global strategy in which all the local information are integrated to yield detailed instruction descript; To achieve reasoning, we designed multiple-turn QA pair to provide the reasoning ability for a model. The empirical results show that the fine-tuned MLLMs by RS-GPT4V can describe fine-grained information. The dataset is available at: https://github.com/GeoX-Lab/RS-GPT4V.
1309.0869
EPTCS
Thao Dang (CNRS-VERIMAG), Tommaso Dreossi (VERIMAG, University of Udine)
Falsifying Oscillation Properties of Parametric Biological Models
In Proceedings HSB 2013, arXiv:1308.5724
EPTCS 125, 2013, pp. 53-67
10.4204/EPTCS.125.4
null
cs.LO cs.CE cs.SY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We propose an approach to falsification of oscillation properties of parametric biological models, based on the recently developed techniques for testing continuous and hybrid systems. In this approach, an oscillation property can be specified using a hybrid automaton, which is then used to guide the exploration in the state and input spaces to search for the behaviors that do not satisfy the property. We illustrate the approach with the Laub-Loomis model for spontaneous oscillations during the aggregation stage of Dictyostelium.
[ { "created": "Tue, 3 Sep 2013 23:41:23 GMT", "version": "v1" } ]
2013-09-05
[ [ "Dang", "Thao", "", "CNRS-VERIMAG" ], [ "Dreossi", "Tommaso", "", "VERIMAG, University of\n Udine" ] ]
We propose an approach to falsification of oscillation properties of parametric biological models, based on the recently developed techniques for testing continuous and hybrid systems. In this approach, an oscillation property can be specified using a hybrid automaton, which is then used to guide the exploration in the state and input spaces to search for the behaviors that do not satisfy the property. We illustrate the approach with the Laub-Loomis model for spontaneous oscillations during the aggregation stage of Dictyostelium.
2210.03026
Germ\'an Vidal
Germ\'an Vidal
Computing Race Variants in Message-Passing Concurrent Programming with Selective Receives
Published as: Vidal, G. (2022). Computing Race Variants in Message-Passing Concurrent Programming with Selective Receives. In: Mousavi, M.R., Philippou, A. (eds) FORTE 2022. Lecture Notes in Computer Science, vol 13273. Springer, Cham. The final authenticated publication is available online at https://doi.org/10.1007/978-3-031-08679-3_12. arXiv admin note: text overlap with arXiv:2112.12869
null
10.1007/978-3-031-08679-3_12
null
cs.PL
http://creativecommons.org/licenses/by-nc-nd/4.0/
Message-passing concurrency is a popular computation model that underlies several programming languages like, e.g., Erlang, Akka, and (to some extent) Go and Rust. In particular, we consider a message-passing concurrent language with dynamic process spawning and selective receives, i.e., where messages can only be consumed by the target process when they match a specific constraint (e.g., the case of Erlang). In this work, we introduce a notion of trace that can be seen as an abstraction of a class of causally equivalent executions (i.e., which produce the same outcome). We then show that execution traces can be used to identify message races. We provide constructive definitions to compute message races as well as to produce so-called race variants, which can then be used to drive new executions which are not causally equivalent to the previous ones. This is an essential ingredient of state-space exploration techniques for program verification.
[ { "created": "Thu, 6 Oct 2022 16:19:15 GMT", "version": "v1" } ]
2022-10-07
[ [ "Vidal", "Germán", "" ] ]
Message-passing concurrency is a popular computation model that underlies several programming languages like, e.g., Erlang, Akka, and (to some extent) Go and Rust. In particular, we consider a message-passing concurrent language with dynamic process spawning and selective receives, i.e., where messages can only be consumed by the target process when they match a specific constraint (e.g., the case of Erlang). In this work, we introduce a notion of trace that can be seen as an abstraction of a class of causally equivalent executions (i.e., which produce the same outcome). We then show that execution traces can be used to identify message races. We provide constructive definitions to compute message races as well as to produce so-called race variants, which can then be used to drive new executions which are not causally equivalent to the previous ones. This is an essential ingredient of state-space exploration techniques for program verification.
1206.0233
Arne Leitert
Arne Leitert
3-Colourability of Dually Chordal Graphs in Linear Time
null
null
null
null
cs.DM
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
A graph G is dually chordal if there is a spanning tree T of G such that any maximal clique of G induces a subtree in T. This paper investigates the Colourability problem on dually chordal graphs. It will show that it is NP-complete in case of four colours and solvable in linear time with a simple algorithm in case of three colours. In addition, it will be shown that a dually chordal graph is 3-colourable if and only if it is perfect and has no clique of size four.
[ { "created": "Fri, 1 Jun 2012 16:00:59 GMT", "version": "v1" }, { "created": "Sun, 5 Aug 2012 15:23:07 GMT", "version": "v2" }, { "created": "Tue, 13 Nov 2012 15:26:12 GMT", "version": "v3" } ]
2012-11-14
[ [ "Leitert", "Arne", "" ] ]
A graph G is dually chordal if there is a spanning tree T of G such that any maximal clique of G induces a subtree in T. This paper investigates the Colourability problem on dually chordal graphs. It will show that it is NP-complete in case of four colours and solvable in linear time with a simple algorithm in case of three colours. In addition, it will be shown that a dually chordal graph is 3-colourable if and only if it is perfect and has no clique of size four.
1404.0818
Marcin Pilipczuk
Daniel Lokshtanov and Marcin Pilipczuk and Micha{\l} Pilipczuk and Saket Saurabh
Fixed-parameter tractable canonization and isomorphism test for graphs of bounded treewidth
Full version of a paper presented at FOCS 2014
null
null
null
cs.DS cs.CC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We give a fixed-parameter tractable algorithm that, given a parameter $k$ and two graphs $G_1,G_2$, either concludes that one of these graphs has treewidth at least $k$, or determines whether $G_1$ and $G_2$ are isomorphic. The running time of the algorithm on an $n$-vertex graph is $2^{O(k^5\log k)}\cdot n^5$, and this is the first fixed-parameter algorithm for Graph Isomorphism parameterized by treewidth. Our algorithm in fact solves the more general canonization problem. We namely design a procedure working in $2^{O(k^5\log k)}\cdot n^5$ time that, for a given graph $G$ on $n$ vertices, either concludes that the treewidth of $G$ is at least $k$, or: * finds in an isomorphic-invariant way a graph $\mathfrak{c}(G)$ that is isomorphic to $G$; * finds an isomorphism-invariant construction term --- an algebraic expression that encodes $G$ together with a tree decomposition of $G$ of width $O(k^4)$. Hence, the isomorphism test reduces to verifying whether the computed isomorphic copies or the construction terms for $G_1$ and $G_2$ are equal.
[ { "created": "Thu, 3 Apr 2014 09:49:54 GMT", "version": "v1" }, { "created": "Wed, 10 Dec 2014 11:32:25 GMT", "version": "v2" } ]
2014-12-11
[ [ "Lokshtanov", "Daniel", "" ], [ "Pilipczuk", "Marcin", "" ], [ "Pilipczuk", "Michał", "" ], [ "Saurabh", "Saket", "" ] ]
We give a fixed-parameter tractable algorithm that, given a parameter $k$ and two graphs $G_1,G_2$, either concludes that one of these graphs has treewidth at least $k$, or determines whether $G_1$ and $G_2$ are isomorphic. The running time of the algorithm on an $n$-vertex graph is $2^{O(k^5\log k)}\cdot n^5$, and this is the first fixed-parameter algorithm for Graph Isomorphism parameterized by treewidth. Our algorithm in fact solves the more general canonization problem. We namely design a procedure working in $2^{O(k^5\log k)}\cdot n^5$ time that, for a given graph $G$ on $n$ vertices, either concludes that the treewidth of $G$ is at least $k$, or: * finds in an isomorphic-invariant way a graph $\mathfrak{c}(G)$ that is isomorphic to $G$; * finds an isomorphism-invariant construction term --- an algebraic expression that encodes $G$ together with a tree decomposition of $G$ of width $O(k^4)$. Hence, the isomorphism test reduces to verifying whether the computed isomorphic copies or the construction terms for $G_1$ and $G_2$ are equal.
2212.02014
Heng Guo
Heng Guo, Jianfeng Zhang, Ke Yan, Le Lu, Minfeng Xu
Med-Query: Steerable Parsing of 9-DoF Medical Anatomies with Query Embedding
updated version
null
null
null
cs.CV
http://creativecommons.org/licenses/by-sa/4.0/
Automatic parsing of human anatomies at instance-level from 3D computed tomography (CT) scans is a prerequisite step for many clinical applications. The presence of pathologies, broken structures or limited field-of-view (FOV) all can make anatomy parsing algorithms vulnerable. In this work, we explore how to exploit and conduct the prosperous detection-then-segmentation paradigm in 3D medical data, and propose a steerable, robust, and efficient computing framework for detection, identification, and segmentation of anatomies in CT scans. Considering complicated shapes, sizes and orientations of anatomies, without lose of generality, we present the nine degrees-of-freedom (9-DoF) pose estimation solution in full 3D space using a novel single-stage, non-hierarchical forward representation. Our whole framework is executed in a steerable manner where any anatomy of interest can be directly retrieved to further boost the inference efficiency. We have validated the proposed method on three medical imaging parsing tasks of ribs, spine, and abdominal organs. For rib parsing, CT scans have been annotated at the rib instance-level for quantitative evaluation, similarly for spine vertebrae and abdominal organs. Extensive experiments on 9-DoF box detection and rib instance segmentation demonstrate the effectiveness of our framework (with the identification rate of 97.0% and the segmentation Dice score of 90.9%) in high efficiency, compared favorably against several strong baselines (e.g., CenterNet, FCOS, and nnU-Net). For spine identification and segmentation, our method achieves a new state-of-the-art result on the public CTSpine1K dataset. Last, we report highly competitive results in multi-organ segmentation at FLARE22 competition. Our annotations, code and models will be made publicly available at: https://github.com/alibaba-damo-academy/Med_Query.
[ { "created": "Mon, 5 Dec 2022 04:04:21 GMT", "version": "v1" }, { "created": "Tue, 10 Oct 2023 10:03:24 GMT", "version": "v2" } ]
2023-10-11
[ [ "Guo", "Heng", "" ], [ "Zhang", "Jianfeng", "" ], [ "Yan", "Ke", "" ], [ "Lu", "Le", "" ], [ "Xu", "Minfeng", "" ] ]
Automatic parsing of human anatomies at instance-level from 3D computed tomography (CT) scans is a prerequisite step for many clinical applications. The presence of pathologies, broken structures or limited field-of-view (FOV) all can make anatomy parsing algorithms vulnerable. In this work, we explore how to exploit and conduct the prosperous detection-then-segmentation paradigm in 3D medical data, and propose a steerable, robust, and efficient computing framework for detection, identification, and segmentation of anatomies in CT scans. Considering complicated shapes, sizes and orientations of anatomies, without lose of generality, we present the nine degrees-of-freedom (9-DoF) pose estimation solution in full 3D space using a novel single-stage, non-hierarchical forward representation. Our whole framework is executed in a steerable manner where any anatomy of interest can be directly retrieved to further boost the inference efficiency. We have validated the proposed method on three medical imaging parsing tasks of ribs, spine, and abdominal organs. For rib parsing, CT scans have been annotated at the rib instance-level for quantitative evaluation, similarly for spine vertebrae and abdominal organs. Extensive experiments on 9-DoF box detection and rib instance segmentation demonstrate the effectiveness of our framework (with the identification rate of 97.0% and the segmentation Dice score of 90.9%) in high efficiency, compared favorably against several strong baselines (e.g., CenterNet, FCOS, and nnU-Net). For spine identification and segmentation, our method achieves a new state-of-the-art result on the public CTSpine1K dataset. Last, we report highly competitive results in multi-organ segmentation at FLARE22 competition. Our annotations, code and models will be made publicly available at: https://github.com/alibaba-damo-academy/Med_Query.
2303.04901
Laura Zheng
Laura Zheng, Julio Poveda, James Mullen, Shreelekha Revankar, Ming C. Lin
Towards Driving Policies with Personality: Modeling Behavior and Style in Risky Scenarios via Data Collection in Virtual Reality
null
null
null
null
cs.RO cs.HC
http://creativecommons.org/licenses/by-nc-sa/4.0/
Autonomous driving research currently faces data sparsity in representation of risky scenarios. Such data is both difficult to obtain ethically in the real world, and unreliable to obtain via simulation. Recent advances in virtual reality (VR) driving simulators lower barriers to tackling this problem in simulation. We propose the first data collection framework for risky scenario driving data from real humans using VR, as well as accompanying numerical driving personality characterizations. We validate the resulting dataset with statistical analyses and model driving behavior with an eight-factor personality vector based on the Multi-dimensional Driving Style Inventory (MDSI). Our method, dataset, and analyses show that realistic driving personalities can be modeled without deep learning or large datasets to complement autonomous driving research.
[ { "created": "Wed, 8 Mar 2023 21:38:24 GMT", "version": "v1" } ]
2023-03-10
[ [ "Zheng", "Laura", "" ], [ "Poveda", "Julio", "" ], [ "Mullen", "James", "" ], [ "Revankar", "Shreelekha", "" ], [ "Lin", "Ming C.", "" ] ]
Autonomous driving research currently faces data sparsity in representation of risky scenarios. Such data is both difficult to obtain ethically in the real world, and unreliable to obtain via simulation. Recent advances in virtual reality (VR) driving simulators lower barriers to tackling this problem in simulation. We propose the first data collection framework for risky scenario driving data from real humans using VR, as well as accompanying numerical driving personality characterizations. We validate the resulting dataset with statistical analyses and model driving behavior with an eight-factor personality vector based on the Multi-dimensional Driving Style Inventory (MDSI). Our method, dataset, and analyses show that realistic driving personalities can be modeled without deep learning or large datasets to complement autonomous driving research.
1801.01451
Andrew Kiruluta
Andrew Kiruluta
Reducing Deep Network Complexity with Fourier Transform Methods
mistake in tensorflow code with test data leakage into training set leading to model over fitting
null
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We propose a novel way that uses shallow densely connected neuron network architectures to achieve superior performance to convolution based neural networks (CNNs) approaches with the added benefits of lower computation burden requiring dramatically less training examples to achieve high prediction accuracy ($>98\%$). The advantages of our proposed method is demonstrated in results on benchmark datasets which show significant performance gain over existing state-of-the-art results on MNIST, CIFAR-10 and CIFAR-100. By Fourier transforming the inputs, each point in the training sample then has a representational energy of all the weighted information from every other point. The consequence of using this input is a reduced complexity neuron network, reduced computation load and the lifting of the requirement for a large number of training examples to achieve high classification accuracy.
[ { "created": "Fri, 15 Dec 2017 20:30:09 GMT", "version": "v1" }, { "created": "Thu, 7 Jun 2018 12:09:37 GMT", "version": "v2" } ]
2018-06-08
[ [ "Kiruluta", "Andrew", "" ] ]
We propose a novel way that uses shallow densely connected neuron network architectures to achieve superior performance to convolution based neural networks (CNNs) approaches with the added benefits of lower computation burden requiring dramatically less training examples to achieve high prediction accuracy ($>98\%$). The advantages of our proposed method is demonstrated in results on benchmark datasets which show significant performance gain over existing state-of-the-art results on MNIST, CIFAR-10 and CIFAR-100. By Fourier transforming the inputs, each point in the training sample then has a representational energy of all the weighted information from every other point. The consequence of using this input is a reduced complexity neuron network, reduced computation load and the lifting of the requirement for a large number of training examples to achieve high classification accuracy.
2302.01301
Raffaele Galliera
Raffaele Galliera, Alessandro Morelli, Roberto Fronteddu, Niranjan Suri
MARLIN: Soft Actor-Critic based Reinforcement Learning for Congestion Control in Real Networks
10 pages, 5 figures, AAAI 2023 workshop "Reinforcement Learning Ready for Production", accepted at NOMS 2023 - IEEE/IFIP Network Operations and Management Symposium
null
10.1109/NOMS56928.2023.10154210
null
cs.LG cs.AI cs.NI
http://creativecommons.org/licenses/by/4.0/
Fast and efficient transport protocols are the foundation of an increasingly distributed world. The burden of continuously delivering improved communication performance to support next-generation applications and services, combined with the increasing heterogeneity of systems and network technologies, has promoted the design of Congestion Control (CC) algorithms that perform well under specific environments. The challenge of designing a generic CC algorithm that can adapt to a broad range of scenarios is still an open research question. To tackle this challenge, we propose to apply a novel Reinforcement Learning (RL) approach. Our solution, MARLIN, uses the Soft Actor-Critic algorithm to maximize both entropy and return and models the learning process as an infinite-horizon task. We trained MARLIN on a real network with varying background traffic patterns to overcome the sim-to-real mismatch that researchers have encountered when applying RL to CC. We evaluated our solution on the task of file transfer and compared it to TCP Cubic. While further research is required, results have shown that MARLIN can achieve comparable results to TCP with little hyperparameter tuning, in a task significantly different from its training setting. Therefore, we believe that our work represents a promising first step toward building CC algorithms based on the maximum entropy RL framework.
[ { "created": "Thu, 2 Feb 2023 18:27:20 GMT", "version": "v1" } ]
2023-06-27
[ [ "Galliera", "Raffaele", "" ], [ "Morelli", "Alessandro", "" ], [ "Fronteddu", "Roberto", "" ], [ "Suri", "Niranjan", "" ] ]
Fast and efficient transport protocols are the foundation of an increasingly distributed world. The burden of continuously delivering improved communication performance to support next-generation applications and services, combined with the increasing heterogeneity of systems and network technologies, has promoted the design of Congestion Control (CC) algorithms that perform well under specific environments. The challenge of designing a generic CC algorithm that can adapt to a broad range of scenarios is still an open research question. To tackle this challenge, we propose to apply a novel Reinforcement Learning (RL) approach. Our solution, MARLIN, uses the Soft Actor-Critic algorithm to maximize both entropy and return and models the learning process as an infinite-horizon task. We trained MARLIN on a real network with varying background traffic patterns to overcome the sim-to-real mismatch that researchers have encountered when applying RL to CC. We evaluated our solution on the task of file transfer and compared it to TCP Cubic. While further research is required, results have shown that MARLIN can achieve comparable results to TCP with little hyperparameter tuning, in a task significantly different from its training setting. Therefore, we believe that our work represents a promising first step toward building CC algorithms based on the maximum entropy RL framework.
2006.14859
Bruno Lecouat
Bruno Lecouat, Jean Ponce, Julien Mairal
A Flexible Framework for Designing Trainable Priors with Adaptive Smoothing and Game Encoding
NeurIPS 2020
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We introduce a general framework for designing and training neural network layers whose forward passes can be interpreted as solving non-smooth convex optimization problems, and whose architectures are derived from an optimization algorithm. We focus on convex games, solved by local agents represented by the nodes of a graph and interacting through regularization functions. This approach is appealing for solving imaging problems, as it allows the use of classical image priors within deep models that are trainable end to end. The priors used in this presentation include variants of total variation, Laplacian regularization, bilateral filtering, sparse coding on learned dictionaries, and non-local self similarities. Our models are fully interpretable as well as parameter and data efficient. Our experiments demonstrate their effectiveness on a large diversity of tasks ranging from image denoising and compressed sensing for fMRI to dense stereo matching.
[ { "created": "Fri, 26 Jun 2020 08:34:54 GMT", "version": "v1" }, { "created": "Mon, 9 Nov 2020 10:00:10 GMT", "version": "v2" } ]
2020-11-10
[ [ "Lecouat", "Bruno", "" ], [ "Ponce", "Jean", "" ], [ "Mairal", "Julien", "" ] ]
We introduce a general framework for designing and training neural network layers whose forward passes can be interpreted as solving non-smooth convex optimization problems, and whose architectures are derived from an optimization algorithm. We focus on convex games, solved by local agents represented by the nodes of a graph and interacting through regularization functions. This approach is appealing for solving imaging problems, as it allows the use of classical image priors within deep models that are trainable end to end. The priors used in this presentation include variants of total variation, Laplacian regularization, bilateral filtering, sparse coding on learned dictionaries, and non-local self similarities. Our models are fully interpretable as well as parameter and data efficient. Our experiments demonstrate their effectiveness on a large diversity of tasks ranging from image denoising and compressed sensing for fMRI to dense stereo matching.
1903.06800
Alessandro Betti
Lorenzo Gigoni, Alessandro Betti, Emanuele Crisostomi, Alessandro Franco, Mauro Tucci, Fabrizio Bizzarri, Debora Mucci
Day-Ahead Hourly Forecasting of Power Generation from Photovoltaic Plants
Preprint of IEEE Transactions of Sustainable Energy, Vol. 9, Issue 2, pp. 831 - 842 (2018)
IEEE Transactions of Sustainable Energy, Vol. 9, Issue 2, pp. 831 - 842 (2018)
10.1109/TSTE.2017.2762435
null
cs.LG stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The ability to accurately forecast power generation from renewable sources is nowadays recognised as a fundamental skill to improve the operation of power systems. Despite the general interest of the power community in this topic, it is not always simple to compare different forecasting methodologies, and infer the impact of single components in providing accurate predictions. In this paper we extensively compare simple forecasting methodologies with more sophisticated ones over 32 photovoltaic plants of different size and technology over a whole year. Also, we try to evaluate the impact of weather conditions and weather forecasts on the prediction of PV power generation.
[ { "created": "Tue, 26 Feb 2019 11:29:18 GMT", "version": "v1" } ]
2019-03-19
[ [ "Gigoni", "Lorenzo", "" ], [ "Betti", "Alessandro", "" ], [ "Crisostomi", "Emanuele", "" ], [ "Franco", "Alessandro", "" ], [ "Tucci", "Mauro", "" ], [ "Bizzarri", "Fabrizio", "" ], [ "Mucci", "Debora", "" ] ]
The ability to accurately forecast power generation from renewable sources is nowadays recognised as a fundamental skill to improve the operation of power systems. Despite the general interest of the power community in this topic, it is not always simple to compare different forecasting methodologies, and infer the impact of single components in providing accurate predictions. In this paper we extensively compare simple forecasting methodologies with more sophisticated ones over 32 photovoltaic plants of different size and technology over a whole year. Also, we try to evaluate the impact of weather conditions and weather forecasts on the prediction of PV power generation.
1806.06397
Karim Armanious
Karim Armanious, Chenming Jiang, Marc Fischer, Thomas K\"ustner, Konstantin Nikolaou, Sergios Gatidis, Bin Yang
MedGAN: Medical Image Translation using GANs
16 pages, 8 figures
null
10.1016/j.compmedimag.2019.101684
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Image-to-image translation is considered a new frontier in the field of medical image analysis, with numerous potential applications. However, a large portion of recent approaches offers individualized solutions based on specialized task-specific architectures or require refinement through non-end-to-end training. In this paper, we propose a new framework, named MedGAN, for medical image-to-image translation which operates on the image level in an end-to-end manner. MedGAN builds upon recent advances in the field of generative adversarial networks (GANs) by merging the adversarial framework with a new combination of non-adversarial losses. We utilize a discriminator network as a trainable feature extractor which penalizes the discrepancy between the translated medical images and the desired modalities. Moreover, style-transfer losses are utilized to match the textures and fine-structures of the desired target images to the translated images. Additionally, we present a new generator architecture, titled CasNet, which enhances the sharpness of the translated medical outputs through progressive refinement via encoder-decoder pairs. Without any application-specific modifications, we apply MedGAN on three different tasks: PET-CT translation, correction of MR motion artefacts and PET image denoising. Perceptual analysis by radiologists and quantitative evaluations illustrate that the MedGAN outperforms other existing translation approaches.
[ { "created": "Sun, 17 Jun 2018 15:45:10 GMT", "version": "v1" }, { "created": "Thu, 4 Apr 2019 14:34:21 GMT", "version": "v2" } ]
2019-11-26
[ [ "Armanious", "Karim", "" ], [ "Jiang", "Chenming", "" ], [ "Fischer", "Marc", "" ], [ "Küstner", "Thomas", "" ], [ "Nikolaou", "Konstantin", "" ], [ "Gatidis", "Sergios", "" ], [ "Yang", "Bin", "" ] ]
Image-to-image translation is considered a new frontier in the field of medical image analysis, with numerous potential applications. However, a large portion of recent approaches offers individualized solutions based on specialized task-specific architectures or require refinement through non-end-to-end training. In this paper, we propose a new framework, named MedGAN, for medical image-to-image translation which operates on the image level in an end-to-end manner. MedGAN builds upon recent advances in the field of generative adversarial networks (GANs) by merging the adversarial framework with a new combination of non-adversarial losses. We utilize a discriminator network as a trainable feature extractor which penalizes the discrepancy between the translated medical images and the desired modalities. Moreover, style-transfer losses are utilized to match the textures and fine-structures of the desired target images to the translated images. Additionally, we present a new generator architecture, titled CasNet, which enhances the sharpness of the translated medical outputs through progressive refinement via encoder-decoder pairs. Without any application-specific modifications, we apply MedGAN on three different tasks: PET-CT translation, correction of MR motion artefacts and PET image denoising. Perceptual analysis by radiologists and quantitative evaluations illustrate that the MedGAN outperforms other existing translation approaches.
1811.00472
Weidi Xie
Erika Lu, Weidi Xie and Andrew Zisserman
Class-Agnostic Counting
Asian Conference on Computer Vision (ACCV), 2018
null
null
null
cs.CV cs.LG
http://creativecommons.org/licenses/by/4.0/
Nearly all existing counting methods are designed for a specific object class. Our work, however, aims to create a counting model able to count any class of object. To achieve this goal, we formulate counting as a matching problem, enabling us to exploit the image self-similarity property that naturally exists in object counting problems. We make the following three contributions: first, a Generic Matching Network (GMN) architecture that can potentially count any object in a class-agnostic manner; second, by reformulating the counting problem as one of matching objects, we can take advantage of the abundance of video data labeled for tracking, which contains natural repetitions suitable for training a counting model. Such data enables us to train the GMN. Third, to customize the GMN to different user requirements, an adapter module is used to specialize the model with minimal effort, i.e. using a few labeled examples, and adapting only a small fraction of the trained parameters. This is a form of few-shot learning, which is practical for domains where labels are limited due to requiring expert knowledge (e.g. microbiology). We demonstrate the flexibility of our method on a diverse set of existing counting benchmarks: specifically cells, cars, and human crowds. The model achieves competitive performance on cell and crowd counting datasets, and surpasses the state-of-the-art on the car dataset using only three training images. When training on the entire dataset, the proposed method outperforms all previous methods by a large margin.
[ { "created": "Thu, 1 Nov 2018 16:11:42 GMT", "version": "v1" } ]
2018-11-02
[ [ "Lu", "Erika", "" ], [ "Xie", "Weidi", "" ], [ "Zisserman", "Andrew", "" ] ]
Nearly all existing counting methods are designed for a specific object class. Our work, however, aims to create a counting model able to count any class of object. To achieve this goal, we formulate counting as a matching problem, enabling us to exploit the image self-similarity property that naturally exists in object counting problems. We make the following three contributions: first, a Generic Matching Network (GMN) architecture that can potentially count any object in a class-agnostic manner; second, by reformulating the counting problem as one of matching objects, we can take advantage of the abundance of video data labeled for tracking, which contains natural repetitions suitable for training a counting model. Such data enables us to train the GMN. Third, to customize the GMN to different user requirements, an adapter module is used to specialize the model with minimal effort, i.e. using a few labeled examples, and adapting only a small fraction of the trained parameters. This is a form of few-shot learning, which is practical for domains where labels are limited due to requiring expert knowledge (e.g. microbiology). We demonstrate the flexibility of our method on a diverse set of existing counting benchmarks: specifically cells, cars, and human crowds. The model achieves competitive performance on cell and crowd counting datasets, and surpasses the state-of-the-art on the car dataset using only three training images. When training on the entire dataset, the proposed method outperforms all previous methods by a large margin.
1810.02490
Mostafa Zaman Chowdhury
Mostafa Zaman Chowdhury and Yeong Min Jang
CAC and Traffic Modeling for Integrated Macrocell/Femtocell Networks
International Conference on Ubiquitous and Future Networks (ICUFN), July 2012, Thailand
null
10.1109/ICUFN.2012.6261709
null
cs.NI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Dense femtocells and the integration of these femtocells with the macrocell are the ultimate goal of the femtocellular network deployment. Integrated macrocell/femtocell networks surely able to provide high data rate for the indoor users as well as able to offload huge traffic from the macrocellular networks to femtocellular networks. Efficient handling of handover calls is the key for the successful macrocell/femtocell integration. An appropriate traffic model for the integrated macrocell/femtocell networks is also needed for the performance analysis measurement. In this paper we presented a call admission control process and a traffic model for the integrated macrocell/femtocell networks. The numerical and simulation results show the important of the integrated macrocell/femtocell network and the performance improvement of the proposed schemes.
[ { "created": "Fri, 5 Oct 2018 01:57:36 GMT", "version": "v1" } ]
2018-10-08
[ [ "Chowdhury", "Mostafa Zaman", "" ], [ "Jang", "Yeong Min", "" ] ]
Dense femtocells and the integration of these femtocells with the macrocell are the ultimate goal of the femtocellular network deployment. Integrated macrocell/femtocell networks surely able to provide high data rate for the indoor users as well as able to offload huge traffic from the macrocellular networks to femtocellular networks. Efficient handling of handover calls is the key for the successful macrocell/femtocell integration. An appropriate traffic model for the integrated macrocell/femtocell networks is also needed for the performance analysis measurement. In this paper we presented a call admission control process and a traffic model for the integrated macrocell/femtocell networks. The numerical and simulation results show the important of the integrated macrocell/femtocell network and the performance improvement of the proposed schemes.
2103.04904
Laszlo Csirmaz
Laszlo Csirmaz, Franti\v{s}ek Mat\'u\v{s} and Carles Padr\'o
Bipartite secret sharing and staircases
To appear in Discrete Mathematics
null
null
null
cs.CR cs.IT math.IT
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Bipartite secret sharing schemes have a bipartite access structure in which the set of participants is divided into two parts and all participants in the same part play an equivalent role. Such a bipartite scheme can be described by a \emph{staircase}: the collection of its minimal points. The complexity of a scheme is the maximal share size relative to the secret size; and the $\kappa$-complexity of an access structure is the best lower bound provided by the entropy method. An access structure is $\kappa$-ideal if it has $\kappa$-complexity 1. Motivated by the abundance of open problems in this area, the main results can be summarized as follows. First, a new characterization of $\kappa$-ideal multipartite access structures is given which offers a straightforward and simple approach to describe ideal bipartite and tripartite access structures. Second, the $\kappa$-complexity is determined for a range of bipartite access structures, including those determined by two points, staircases with equal widths and heights, and staircases with all heights 1. Third, matching linear schemes are presented for some non-ideal cases, including staircases where all heights are 1 and all widths are equal. Finally, finding the Shannon complexity of a bipartite access structure can be considered as a discrete submodular optimization problem. An interesting and intriguing continuous version is defined which might give further insight to the large-scale behavior of these optimization problems.
[ { "created": "Mon, 8 Mar 2021 17:09:43 GMT", "version": "v1" }, { "created": "Thu, 5 Oct 2023 14:19:21 GMT", "version": "v2" } ]
2023-10-06
[ [ "Csirmaz", "Laszlo", "" ], [ "Matúš", "František", "" ], [ "Padró", "Carles", "" ] ]
Bipartite secret sharing schemes have a bipartite access structure in which the set of participants is divided into two parts and all participants in the same part play an equivalent role. Such a bipartite scheme can be described by a \emph{staircase}: the collection of its minimal points. The complexity of a scheme is the maximal share size relative to the secret size; and the $\kappa$-complexity of an access structure is the best lower bound provided by the entropy method. An access structure is $\kappa$-ideal if it has $\kappa$-complexity 1. Motivated by the abundance of open problems in this area, the main results can be summarized as follows. First, a new characterization of $\kappa$-ideal multipartite access structures is given which offers a straightforward and simple approach to describe ideal bipartite and tripartite access structures. Second, the $\kappa$-complexity is determined for a range of bipartite access structures, including those determined by two points, staircases with equal widths and heights, and staircases with all heights 1. Third, matching linear schemes are presented for some non-ideal cases, including staircases where all heights are 1 and all widths are equal. Finally, finding the Shannon complexity of a bipartite access structure can be considered as a discrete submodular optimization problem. An interesting and intriguing continuous version is defined which might give further insight to the large-scale behavior of these optimization problems.
2103.02649
Xiaoyang Wang
Xiaoyang Wang, Jonathan D Thomas, Robert J Piechocki, Shipra Kapoor, Raul Santos-Rodriguez, Arjun Parekh
Self-play Learning Strategies for Resource Assignment in Open-RAN Networks
null
null
null
null
cs.NI cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Open Radio Access Network (ORAN) is being developed with an aim to democratise access and lower the cost of future mobile data networks, supporting network services with various QoS requirements, such as massive IoT and URLLC. In ORAN, network functionality is dis-aggregated into remote units (RUs), distributed units (DUs) and central units (CUs), which allows flexible software on Commercial-Off-The-Shelf (COTS) deployments. Furthermore, the mapping of variable RU requirements to local mobile edge computing centres for future centralized processing would significantly reduce the power consumption in cellular networks. In this paper, we study the RU-DU resource assignment problem in an ORAN system, modelled as a 2D bin packing problem. A deep reinforcement learning-based self-play approach is proposed to achieve efficient RU-DU resource management, with AlphaGo Zero inspired neural Monte-Carlo Tree Search (MCTS). Experiments on representative 2D bin packing environment and real sites data show that the self-play learning strategy achieves intelligent RU-DU resource assignment for different network conditions.
[ { "created": "Wed, 3 Mar 2021 19:31:29 GMT", "version": "v1" } ]
2021-03-05
[ [ "Wang", "Xiaoyang", "" ], [ "Thomas", "Jonathan D", "" ], [ "Piechocki", "Robert J", "" ], [ "Kapoor", "Shipra", "" ], [ "Santos-Rodriguez", "Raul", "" ], [ "Parekh", "Arjun", "" ] ]
Open Radio Access Network (ORAN) is being developed with an aim to democratise access and lower the cost of future mobile data networks, supporting network services with various QoS requirements, such as massive IoT and URLLC. In ORAN, network functionality is dis-aggregated into remote units (RUs), distributed units (DUs) and central units (CUs), which allows flexible software on Commercial-Off-The-Shelf (COTS) deployments. Furthermore, the mapping of variable RU requirements to local mobile edge computing centres for future centralized processing would significantly reduce the power consumption in cellular networks. In this paper, we study the RU-DU resource assignment problem in an ORAN system, modelled as a 2D bin packing problem. A deep reinforcement learning-based self-play approach is proposed to achieve efficient RU-DU resource management, with AlphaGo Zero inspired neural Monte-Carlo Tree Search (MCTS). Experiments on representative 2D bin packing environment and real sites data show that the self-play learning strategy achieves intelligent RU-DU resource assignment for different network conditions.
2305.10459
Hadjer Benmeziane
Hadjer Benmeziane, Corey Lammie, Irem Boybat, Malte Rasch, Manuel Le Gallo, Hsinyu Tsai, Ramachandran Muralidhar, Smail Niar, Ouarnoughi Hamza, Vijay Narayanan, Abu Sebastian and Kaoutar El Maghraoui
AnalogNAS: A Neural Network Design Framework for Accurate Inference with Analog In-Memory Computing
Accepted to IEEE Edge
null
null
null
cs.AR cs.CV cs.LG
http://creativecommons.org/licenses/by/4.0/
The advancement of Deep Learning (DL) is driven by efficient Deep Neural Network (DNN) design and new hardware accelerators. Current DNN design is primarily tailored for general-purpose use and deployment on commercially viable platforms. Inference at the edge requires low latency, compact and power-efficient models, and must be cost-effective. Digital processors based on typical von Neumann architectures are not conducive to edge AI given the large amounts of required data movement in and out of memory. Conversely, analog/mixed signal in-memory computing hardware accelerators can easily transcend the memory wall of von Neuman architectures when accelerating inference workloads. They offer increased area and power efficiency, which are paramount in edge resource-constrained environments. In this paper, we propose AnalogNAS, a framework for automated DNN design targeting deployment on analog In-Memory Computing (IMC) inference accelerators. We conduct extensive hardware simulations to demonstrate the performance of AnalogNAS on State-Of-The-Art (SOTA) models in terms of accuracy and deployment efficiency on various Tiny Machine Learning (TinyML) tasks. We also present experimental results that show AnalogNAS models achieving higher accuracy than SOTA models when implemented on a 64-core IMC chip based on Phase Change Memory (PCM). The AnalogNAS search code is released: https://github.com/IBM/analog-nas
[ { "created": "Wed, 17 May 2023 07:39:14 GMT", "version": "v1" } ]
2023-05-19
[ [ "Benmeziane", "Hadjer", "" ], [ "Lammie", "Corey", "" ], [ "Boybat", "Irem", "" ], [ "Rasch", "Malte", "" ], [ "Gallo", "Manuel Le", "" ], [ "Tsai", "Hsinyu", "" ], [ "Muralidhar", "Ramachandran", "" ], [ "Niar", "Smail", "" ], [ "Hamza", "Ouarnoughi", "" ], [ "Narayanan", "Vijay", "" ], [ "Sebastian", "Abu", "" ], [ "Maghraoui", "Kaoutar El", "" ] ]
The advancement of Deep Learning (DL) is driven by efficient Deep Neural Network (DNN) design and new hardware accelerators. Current DNN design is primarily tailored for general-purpose use and deployment on commercially viable platforms. Inference at the edge requires low latency, compact and power-efficient models, and must be cost-effective. Digital processors based on typical von Neumann architectures are not conducive to edge AI given the large amounts of required data movement in and out of memory. Conversely, analog/mixed signal in-memory computing hardware accelerators can easily transcend the memory wall of von Neuman architectures when accelerating inference workloads. They offer increased area and power efficiency, which are paramount in edge resource-constrained environments. In this paper, we propose AnalogNAS, a framework for automated DNN design targeting deployment on analog In-Memory Computing (IMC) inference accelerators. We conduct extensive hardware simulations to demonstrate the performance of AnalogNAS on State-Of-The-Art (SOTA) models in terms of accuracy and deployment efficiency on various Tiny Machine Learning (TinyML) tasks. We also present experimental results that show AnalogNAS models achieving higher accuracy than SOTA models when implemented on a 64-core IMC chip based on Phase Change Memory (PCM). The AnalogNAS search code is released: https://github.com/IBM/analog-nas
2404.05502
Roman Kazakov
Roman Kazakov, Kseniia Petukhova, Ekaterina Kochmar
PetKaz at SemEval-2024 Task 3: Advancing Emotion Classification with an LLM for Emotion-Cause Pair Extraction in Conversations
8 pages, 7 figures, 2 tables, to be published in the Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024), for associated code, see https://github.com/sachertort/petkaz-semeval-ecac
null
null
null
cs.CL cs.AI
http://creativecommons.org/licenses/by/4.0/
In this paper, we present our submission to the SemEval-2023 Task~3 "The Competition of Multimodal Emotion Cause Analysis in Conversations", focusing on extracting emotion-cause pairs from dialogs. Specifically, our approach relies on combining fine-tuned GPT-3.5 for emotion classification and a BiLSTM-based neural network to detect causes. We score 2nd in the ranking for Subtask 1, demonstrating the effectiveness of our approach through one of the highest weighted-average proportional F1 scores recorded at 0.264.
[ { "created": "Mon, 8 Apr 2024 13:25:03 GMT", "version": "v1" } ]
2024-04-09
[ [ "Kazakov", "Roman", "" ], [ "Petukhova", "Kseniia", "" ], [ "Kochmar", "Ekaterina", "" ] ]
In this paper, we present our submission to the SemEval-2023 Task~3 "The Competition of Multimodal Emotion Cause Analysis in Conversations", focusing on extracting emotion-cause pairs from dialogs. Specifically, our approach relies on combining fine-tuned GPT-3.5 for emotion classification and a BiLSTM-based neural network to detect causes. We score 2nd in the ranking for Subtask 1, demonstrating the effectiveness of our approach through one of the highest weighted-average proportional F1 scores recorded at 0.264.
2306.07042
Enric Boix-Adser\`a
Enric Boix-Adsera, Etai Littwin, Emmanuel Abbe, Samy Bengio, Joshua Susskind
Transformers learn through gradual rank increase
39 pages, to appear in NeurIPS 2023
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We identify incremental learning dynamics in transformers, where the difference between trained and initial weights progressively increases in rank. We rigorously prove this occurs under the simplifying assumptions of diagonal weight matrices and small initialization. Our experiments support the theory and also show that phenomenon can occur in practice without the simplifying assumptions.
[ { "created": "Mon, 12 Jun 2023 11:41:42 GMT", "version": "v1" }, { "created": "Mon, 11 Dec 2023 00:23:45 GMT", "version": "v2" } ]
2023-12-12
[ [ "Boix-Adsera", "Enric", "" ], [ "Littwin", "Etai", "" ], [ "Abbe", "Emmanuel", "" ], [ "Bengio", "Samy", "" ], [ "Susskind", "Joshua", "" ] ]
We identify incremental learning dynamics in transformers, where the difference between trained and initial weights progressively increases in rank. We rigorously prove this occurs under the simplifying assumptions of diagonal weight matrices and small initialization. Our experiments support the theory and also show that phenomenon can occur in practice without the simplifying assumptions.
2203.12328
Ananthanarayanan Chockalingam
Sandesh Rao Mattu and A. Chockalingam
Learning based Channel Estimation and Phase Noise Compensation in Doubly-Selective Channels
Comm. Lett. Copyright IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
null
null
null
cs.IT eess.SP math.IT
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this letter, we propose a learning based channel estimation scheme for orthogonal frequency division multiplexing (OFDM) systems in the presence of phase noise in doubly-selective fading channels. Two-dimensional (2D) convolutional neural networks (CNNs) are employed for effective training and tracking of channel variation in both frequency as well as time domain. The proposed network learns and estimates the channel coefficients in the entire time-frequency (TF) grid based on pilots sparsely populated in the TF grid. In order to make the network robust to phase noise (PN) impairment, a novel training scheme where the training data is rotated by random phases before being fed to the network is employed. Further, using the estimated channel coefficients, a simple and effective PN estimation and compensation scheme is devised. Numerical results demonstrate that the proposed network and PN compensation scheme achieve robust OFDM performance in the presence of phase noise.
[ { "created": "Wed, 23 Mar 2022 11:13:27 GMT", "version": "v1" } ]
2022-03-24
[ [ "Mattu", "Sandesh Rao", "" ], [ "Chockalingam", "A.", "" ] ]
In this letter, we propose a learning based channel estimation scheme for orthogonal frequency division multiplexing (OFDM) systems in the presence of phase noise in doubly-selective fading channels. Two-dimensional (2D) convolutional neural networks (CNNs) are employed for effective training and tracking of channel variation in both frequency as well as time domain. The proposed network learns and estimates the channel coefficients in the entire time-frequency (TF) grid based on pilots sparsely populated in the TF grid. In order to make the network robust to phase noise (PN) impairment, a novel training scheme where the training data is rotated by random phases before being fed to the network is employed. Further, using the estimated channel coefficients, a simple and effective PN estimation and compensation scheme is devised. Numerical results demonstrate that the proposed network and PN compensation scheme achieve robust OFDM performance in the presence of phase noise.
2205.14122
Alexandra Fedorova
Alexandra Fedorova, Keith Smith, Keith Bostic, Alexander Gorrod, Sue LoVerso, Michael Cahill
Writes Hurt: Lessons in Cache Design for Optane NVRAM
null
null
null
null
cs.AR cs.DB
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Intel OptaneTM DC Persistent Memory resides on the memory bus and approaches DRAM in access latency. One avenue for its adoption is to employ it in place of persistent storage; another is to use it as a cheaper and denser extension of DRAM. In pursuit of the latter goal, we present the design of a volatile Optane NVRAM cache as a component in a storage engine underlying MongoDB. The primary innovation in our design is a new cache admission policy. We discover that on Optane NVRAM, known for its limited write throughput, the presence of writes disproportionately affects the throughput of reads, much more so than on DRAM. Therefore, an admission policy that indiscriminately admits new data (and thus generates writes), severely limits the rate of data retrieval and results in exceedingly poor performance for the cache overall. We design an admission policy that balances the rate of admission with the rate of lookups using dynamically observed characteristics of the workload. Our implementation outperforms OpenCAS (an off-the-shelf Optane-based block cache) in all cases, and Intel Memory Mode in cases where the database size exceeds the available NVRAM. Our cache is decoupled from the rest of the storage engine and uses generic metrics to guide its admission policy; therefore our design can be easily adopted in other systems.
[ { "created": "Tue, 24 May 2022 22:16:11 GMT", "version": "v1" } ]
2022-05-30
[ [ "Fedorova", "Alexandra", "" ], [ "Smith", "Keith", "" ], [ "Bostic", "Keith", "" ], [ "Gorrod", "Alexander", "" ], [ "LoVerso", "Sue", "" ], [ "Cahill", "Michael", "" ] ]
Intel OptaneTM DC Persistent Memory resides on the memory bus and approaches DRAM in access latency. One avenue for its adoption is to employ it in place of persistent storage; another is to use it as a cheaper and denser extension of DRAM. In pursuit of the latter goal, we present the design of a volatile Optane NVRAM cache as a component in a storage engine underlying MongoDB. The primary innovation in our design is a new cache admission policy. We discover that on Optane NVRAM, known for its limited write throughput, the presence of writes disproportionately affects the throughput of reads, much more so than on DRAM. Therefore, an admission policy that indiscriminately admits new data (and thus generates writes), severely limits the rate of data retrieval and results in exceedingly poor performance for the cache overall. We design an admission policy that balances the rate of admission with the rate of lookups using dynamically observed characteristics of the workload. Our implementation outperforms OpenCAS (an off-the-shelf Optane-based block cache) in all cases, and Intel Memory Mode in cases where the database size exceeds the available NVRAM. Our cache is decoupled from the rest of the storage engine and uses generic metrics to guide its admission policy; therefore our design can be easily adopted in other systems.
2303.13367
Brady Lund
Brady Lund, Ting Wang, Nishith Reddy Mannuru, Bing Nie, Somipam Shimray, and Ziang Wang
ChatGPT and a New Academic Reality: Artificial Intelligence-Written Research Papers and the Ethics of the Large Language Models in Scholarly Publishing
null
Journal of the Association for Information Science and Technology (2023)
10.1002/asi.24750
null
cs.CL cs.CY
http://creativecommons.org/licenses/by/4.0/
This paper discusses OpenAIs ChatGPT, a generative pre-trained transformer, which uses natural language processing to fulfill text-based user requests (i.e., a chatbot). The history and principles behind ChatGPT and similar models are discussed. This technology is then discussed in relation to its potential impact on academia and scholarly research and publishing. ChatGPT is seen as a potential model for the automated preparation of essays and other types of scholarly manuscripts. Potential ethical issues that could arise with the emergence of large language models like GPT-3, the underlying technology behind ChatGPT, and its usage by academics and researchers, are discussed and situated within the context of broader advancements in artificial intelligence, machine learning, and natural language processing for research and scholarly publishing.
[ { "created": "Tue, 21 Mar 2023 14:35:07 GMT", "version": "v1" }, { "created": "Fri, 31 Mar 2023 17:56:28 GMT", "version": "v2" } ]
2023-04-03
[ [ "Lund", "Brady", "" ], [ "Wang", "Ting", "" ], [ "Mannuru", "Nishith Reddy", "" ], [ "Nie", "Bing", "" ], [ "Shimray", "Somipam", "" ], [ "Wang", "Ziang", "" ] ]
This paper discusses OpenAIs ChatGPT, a generative pre-trained transformer, which uses natural language processing to fulfill text-based user requests (i.e., a chatbot). The history and principles behind ChatGPT and similar models are discussed. This technology is then discussed in relation to its potential impact on academia and scholarly research and publishing. ChatGPT is seen as a potential model for the automated preparation of essays and other types of scholarly manuscripts. Potential ethical issues that could arise with the emergence of large language models like GPT-3, the underlying technology behind ChatGPT, and its usage by academics and researchers, are discussed and situated within the context of broader advancements in artificial intelligence, machine learning, and natural language processing for research and scholarly publishing.
2310.20587
Ruizhe Shi
Ruizhe Shi, Yuyao Liu, Yanjie Ze, Simon S. Du, Huazhe Xu
Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning
24 pages, 16 tables
null
null
null
cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Offline reinforcement learning (RL) aims to find a near-optimal policy using pre-collected datasets. In real-world scenarios, data collection could be costly and risky; therefore, offline RL becomes particularly challenging when the in-domain data is limited. Given recent advances in Large Language Models (LLMs) and their few-shot learning prowess, this paper introduces $\textbf{La}$nguage Models for $\textbf{Mo}$tion Control ($\textbf{LaMo}$), a general framework based on Decision Transformers to effectively use pre-trained Language Models (LMs) for offline RL. Our framework highlights four crucial components: (1) Initializing Decision Transformers with sequentially pre-trained LMs, (2) employing the LoRA fine-tuning method, in contrast to full-weight fine-tuning, to combine the pre-trained knowledge from LMs and in-domain knowledge effectively, (3) using the non-linear MLP transformation instead of linear projections, to generate embeddings, and (4) integrating an auxiliary language prediction loss during fine-tuning to stabilize the LMs and retain their original abilities on languages. Empirical results indicate $\textbf{LaMo}$ achieves state-of-the-art performance in sparse-reward tasks and closes the gap between value-based offline RL methods and decision transformers in dense-reward tasks. In particular, our method demonstrates superior performance in scenarios with limited data samples.
[ { "created": "Tue, 31 Oct 2023 16:24:17 GMT", "version": "v1" }, { "created": "Sat, 4 Nov 2023 08:11:10 GMT", "version": "v2" }, { "created": "Tue, 7 Nov 2023 03:26:51 GMT", "version": "v3" }, { "created": "Mon, 27 Nov 2023 07:38:06 GMT", "version": "v4" } ]
2023-11-28
[ [ "Shi", "Ruizhe", "" ], [ "Liu", "Yuyao", "" ], [ "Ze", "Yanjie", "" ], [ "Du", "Simon S.", "" ], [ "Xu", "Huazhe", "" ] ]
Offline reinforcement learning (RL) aims to find a near-optimal policy using pre-collected datasets. In real-world scenarios, data collection could be costly and risky; therefore, offline RL becomes particularly challenging when the in-domain data is limited. Given recent advances in Large Language Models (LLMs) and their few-shot learning prowess, this paper introduces $\textbf{La}$nguage Models for $\textbf{Mo}$tion Control ($\textbf{LaMo}$), a general framework based on Decision Transformers to effectively use pre-trained Language Models (LMs) for offline RL. Our framework highlights four crucial components: (1) Initializing Decision Transformers with sequentially pre-trained LMs, (2) employing the LoRA fine-tuning method, in contrast to full-weight fine-tuning, to combine the pre-trained knowledge from LMs and in-domain knowledge effectively, (3) using the non-linear MLP transformation instead of linear projections, to generate embeddings, and (4) integrating an auxiliary language prediction loss during fine-tuning to stabilize the LMs and retain their original abilities on languages. Empirical results indicate $\textbf{LaMo}$ achieves state-of-the-art performance in sparse-reward tasks and closes the gap between value-based offline RL methods and decision transformers in dense-reward tasks. In particular, our method demonstrates superior performance in scenarios with limited data samples.
2111.09888
Apoorv Khandelwal
Apoorv Khandelwal, Luca Weihs, Roozbeh Mottaghi, Aniruddha Kembhavi
Simple but Effective: CLIP Embeddings for Embodied AI
Published in CVPR 2022
null
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Contrastive language image pretraining (CLIP) encoders have been shown to be beneficial for a range of visual tasks from classification and detection to captioning and image manipulation. We investigate the effectiveness of CLIP visual backbones for Embodied AI tasks. We build incredibly simple baselines, named EmbCLIP, with no task specific architectures, inductive biases (such as the use of semantic maps), auxiliary tasks during training, or depth maps -- yet we find that our improved baselines perform very well across a range of tasks and simulators. EmbCLIP tops the RoboTHOR ObjectNav leaderboard by a huge margin of 20 pts (Success Rate). It tops the iTHOR 1-Phase Rearrangement leaderboard, beating the next best submission, which employs Active Neural Mapping, and more than doubling the % Fixed Strict metric (0.08 to 0.17). It also beats the winners of the 2021 Habitat ObjectNav Challenge, which employ auxiliary tasks, depth maps, and human demonstrations, and those of the 2019 Habitat PointNav Challenge. We evaluate the ability of CLIP's visual representations at capturing semantic information about input observations -- primitives that are useful for navigation-heavy embodied tasks -- and find that CLIP's representations encode these primitives more effectively than ImageNet-pretrained backbones. Finally, we extend one of our baselines, producing an agent capable of zero-shot object navigation that can navigate to objects that were not used as targets during training. Our code and models are available at https://github.com/allenai/embodied-clip
[ { "created": "Thu, 18 Nov 2021 18:59:59 GMT", "version": "v1" }, { "created": "Fri, 15 Apr 2022 02:12:26 GMT", "version": "v2" } ]
2022-04-18
[ [ "Khandelwal", "Apoorv", "" ], [ "Weihs", "Luca", "" ], [ "Mottaghi", "Roozbeh", "" ], [ "Kembhavi", "Aniruddha", "" ] ]
Contrastive language image pretraining (CLIP) encoders have been shown to be beneficial for a range of visual tasks from classification and detection to captioning and image manipulation. We investigate the effectiveness of CLIP visual backbones for Embodied AI tasks. We build incredibly simple baselines, named EmbCLIP, with no task specific architectures, inductive biases (such as the use of semantic maps), auxiliary tasks during training, or depth maps -- yet we find that our improved baselines perform very well across a range of tasks and simulators. EmbCLIP tops the RoboTHOR ObjectNav leaderboard by a huge margin of 20 pts (Success Rate). It tops the iTHOR 1-Phase Rearrangement leaderboard, beating the next best submission, which employs Active Neural Mapping, and more than doubling the % Fixed Strict metric (0.08 to 0.17). It also beats the winners of the 2021 Habitat ObjectNav Challenge, which employ auxiliary tasks, depth maps, and human demonstrations, and those of the 2019 Habitat PointNav Challenge. We evaluate the ability of CLIP's visual representations at capturing semantic information about input observations -- primitives that are useful for navigation-heavy embodied tasks -- and find that CLIP's representations encode these primitives more effectively than ImageNet-pretrained backbones. Finally, we extend one of our baselines, producing an agent capable of zero-shot object navigation that can navigate to objects that were not used as targets during training. Our code and models are available at https://github.com/allenai/embodied-clip
1611.01195
Shusil Dangi
Shusil Dangi, Nathan Cahill, Cristian A. Linte
Integrating Atlas and Graph Cut Methods for LV Segmentation from Cardiac Cine MRI
Statistical Atlases and Computational Modelling of Heart workshop 2016
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Magnetic Resonance Imaging (MRI) has evolved as a clinical standard-of-care imaging modality for cardiac morphology, function assessment, and guidance of cardiac interventions. All these applications rely on accurate extraction of the myocardial tissue and blood pool from the imaging data. Here we propose a framework for left ventricle (LV) segmentation from cardiac cine-MRI. First, we segment the LV blood pool using iterative graph cuts, and subsequently use this information to segment the myocardium. We formulate the segmentation procedure as an energy minimization problem in a graph subject to the shape prior obtained by label propagation from an average atlas using affine registration. The proposed framework has been validated on 30 patient cardiac cine-MRI datasets available through the STACOM LV segmentation challenge and yielded fast, robust, and accurate segmentation results.
[ { "created": "Thu, 3 Nov 2016 21:12:55 GMT", "version": "v1" } ]
2016-11-07
[ [ "Dangi", "Shusil", "" ], [ "Cahill", "Nathan", "" ], [ "Linte", "Cristian A.", "" ] ]
Magnetic Resonance Imaging (MRI) has evolved as a clinical standard-of-care imaging modality for cardiac morphology, function assessment, and guidance of cardiac interventions. All these applications rely on accurate extraction of the myocardial tissue and blood pool from the imaging data. Here we propose a framework for left ventricle (LV) segmentation from cardiac cine-MRI. First, we segment the LV blood pool using iterative graph cuts, and subsequently use this information to segment the myocardium. We formulate the segmentation procedure as an energy minimization problem in a graph subject to the shape prior obtained by label propagation from an average atlas using affine registration. The proposed framework has been validated on 30 patient cardiac cine-MRI datasets available through the STACOM LV segmentation challenge and yielded fast, robust, and accurate segmentation results.
2311.07634
Wenshuai Xu
Wenshuai Xu, Zhenghui Hu, Yu Lu, Jinzhou Meng, Qingjie Liu, Yunhong Wang
ActiveDC: Distribution Calibration for Active Finetuning
CVPR 2024 Accept
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The pretraining-finetuning paradigm has gained popularity in various computer vision tasks. In this paradigm, the emergence of active finetuning arises due to the abundance of large-scale data and costly annotation requirements. Active finetuning involves selecting a subset of data from an unlabeled pool for annotation, facilitating subsequent finetuning. However, the use of a limited number of training samples can lead to a biased distribution, potentially resulting in model overfitting. In this paper, we propose a new method called ActiveDC for the active finetuning tasks. Firstly, we select samples for annotation by optimizing the distribution similarity between the subset to be selected and the entire unlabeled pool in continuous space. Secondly, we calibrate the distribution of the selected samples by exploiting implicit category information in the unlabeled pool. The feature visualization provides an intuitive sense of the effectiveness of our approach to distribution calibration. We conducted extensive experiments on three image classification datasets with different sampling ratios. The results indicate that ActiveDC consistently outperforms the baseline performance in all image classification tasks. The improvement is particularly significant when the sampling ratio is low, with performance gains of up to 10%. Our code will be released.
[ { "created": "Mon, 13 Nov 2023 14:35:18 GMT", "version": "v1" }, { "created": "Wed, 15 Nov 2023 09:32:05 GMT", "version": "v2" }, { "created": "Tue, 27 Feb 2024 07:52:16 GMT", "version": "v3" } ]
2024-02-28
[ [ "Xu", "Wenshuai", "" ], [ "Hu", "Zhenghui", "" ], [ "Lu", "Yu", "" ], [ "Meng", "Jinzhou", "" ], [ "Liu", "Qingjie", "" ], [ "Wang", "Yunhong", "" ] ]
The pretraining-finetuning paradigm has gained popularity in various computer vision tasks. In this paradigm, the emergence of active finetuning arises due to the abundance of large-scale data and costly annotation requirements. Active finetuning involves selecting a subset of data from an unlabeled pool for annotation, facilitating subsequent finetuning. However, the use of a limited number of training samples can lead to a biased distribution, potentially resulting in model overfitting. In this paper, we propose a new method called ActiveDC for the active finetuning tasks. Firstly, we select samples for annotation by optimizing the distribution similarity between the subset to be selected and the entire unlabeled pool in continuous space. Secondly, we calibrate the distribution of the selected samples by exploiting implicit category information in the unlabeled pool. The feature visualization provides an intuitive sense of the effectiveness of our approach to distribution calibration. We conducted extensive experiments on three image classification datasets with different sampling ratios. The results indicate that ActiveDC consistently outperforms the baseline performance in all image classification tasks. The improvement is particularly significant when the sampling ratio is low, with performance gains of up to 10%. Our code will be released.
0803.2812
Mikhail Konnik
Mikhail V. Konnik
Using Spatially Varying Pixels Exposures and Bayer-covered Photosensors for High Dynamic Range Imaging
Typos corrected
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The method of a linear high dynamic range imaging using solid-state photosensors with Bayer colour filters array is provided in this paper. Using information from neighbour pixels, it is possible to reconstruct linear images with wide dynamic range from the oversaturated images. Bayer colour filters array is considered as an array of neutral filters in a quasimonochromatic light. If the camera's response function to the desirable light source is known then one can calculate correction coefficients to reconstruct oversaturated images. Reconstructed images are linearized in order to provide a linear high dynamic range images for optical-digital imaging systems. The calibration procedure for obtaining the camera's response function to the desired light source is described. Experimental results of the reconstruction of the images from the oversaturated images are presented for red, green, and blue quasimonochromatic light sources. Quantitative analysis of the accuracy of the reconstructed images is provided.
[ { "created": "Wed, 19 Mar 2008 14:55:15 GMT", "version": "v1" }, { "created": "Mon, 24 Mar 2008 07:04:39 GMT", "version": "v2" } ]
2008-03-24
[ [ "Konnik", "Mikhail V.", "" ] ]
The method of a linear high dynamic range imaging using solid-state photosensors with Bayer colour filters array is provided in this paper. Using information from neighbour pixels, it is possible to reconstruct linear images with wide dynamic range from the oversaturated images. Bayer colour filters array is considered as an array of neutral filters in a quasimonochromatic light. If the camera's response function to the desirable light source is known then one can calculate correction coefficients to reconstruct oversaturated images. Reconstructed images are linearized in order to provide a linear high dynamic range images for optical-digital imaging systems. The calibration procedure for obtaining the camera's response function to the desired light source is described. Experimental results of the reconstruction of the images from the oversaturated images are presented for red, green, and blue quasimonochromatic light sources. Quantitative analysis of the accuracy of the reconstructed images is provided.
2306.14209
Fabio Merizzi
Fabio Merizzi, Perrine Saillard, Oceane Acquier, Elena Morotti, Elena Loli Piccolomini, Luca Calatroni and Rosa Maria Dess\`i
Deep image prior inpainting of ancient frescoes in the Mediterranean Alpine arc
26 pages
null
null
null
cs.CV cs.AI
http://creativecommons.org/licenses/by-nc-sa/4.0/
The unprecedented success of image reconstruction approaches based on deep neural networks has revolutionised both the processing and the analysis paradigms in several applied disciplines. In the field of digital humanities, the task of digital reconstruction of ancient frescoes is particularly challenging due to the scarce amount of available training data caused by ageing, wear, tear and retouching over time. To overcome these difficulties, we consider the Deep Image Prior (DIP) inpainting approach which computes appropriate reconstructions by relying on the progressive updating of an untrained convolutional neural network so as to match the reliable piece of information in the image at hand while promoting regularisation elsewhere. In comparison with state-of-the-art approaches (based on variational/PDEs and patch-based methods), DIP-based inpainting reduces artefacts and better adapts to contextual/non-local information, thus providing a valuable and effective tool for art historians. As a case study, we apply such approach to reconstruct missing image contents in a dataset of highly damaged digital images of medieval paintings located into several chapels in the Mediterranean Alpine Arc and provide a detailed description on how visible and invisible (e.g., infrared) information can be integrated for identifying and reconstructing damaged image regions.
[ { "created": "Sun, 25 Jun 2023 11:19:47 GMT", "version": "v1" }, { "created": "Mon, 11 Dec 2023 15:30:01 GMT", "version": "v2" } ]
2023-12-14
[ [ "Merizzi", "Fabio", "" ], [ "Saillard", "Perrine", "" ], [ "Acquier", "Oceane", "" ], [ "Morotti", "Elena", "" ], [ "Piccolomini", "Elena Loli", "" ], [ "Calatroni", "Luca", "" ], [ "Dessì", "Rosa Maria", "" ] ]
The unprecedented success of image reconstruction approaches based on deep neural networks has revolutionised both the processing and the analysis paradigms in several applied disciplines. In the field of digital humanities, the task of digital reconstruction of ancient frescoes is particularly challenging due to the scarce amount of available training data caused by ageing, wear, tear and retouching over time. To overcome these difficulties, we consider the Deep Image Prior (DIP) inpainting approach which computes appropriate reconstructions by relying on the progressive updating of an untrained convolutional neural network so as to match the reliable piece of information in the image at hand while promoting regularisation elsewhere. In comparison with state-of-the-art approaches (based on variational/PDEs and patch-based methods), DIP-based inpainting reduces artefacts and better adapts to contextual/non-local information, thus providing a valuable and effective tool for art historians. As a case study, we apply such approach to reconstruct missing image contents in a dataset of highly damaged digital images of medieval paintings located into several chapels in the Mediterranean Alpine Arc and provide a detailed description on how visible and invisible (e.g., infrared) information can be integrated for identifying and reconstructing damaged image regions.
cs/0610151
Anant Sahai
Anant Sahai
Anytime coding on the infinite bandwidth AWGN channel: A sequential semi-orthogonal optimal code
12 pages, 6 figures, submitted to IT Transactions
null
null
null
cs.IT math.IT
null
It is well known that orthogonal coding can be used to approach the Shannon capacity of the power-constrained AWGN channel without a bandwidth constraint. This correspondence describes a semi-orthogonal variation of pulse position modulation that is sequential in nature -- bits can be ``streamed across'' without having to buffer up blocks of bits at the transmitter. ML decoding results in an exponentially small probability of error as a function of tolerated receiver delay and thus eventually a zero probability of error on every transmitted bit. In the high-rate regime, a matching upper bound is given on the delay error exponent. We close with some comments on the case with feedback and the connections to the capacity per unit cost problem.
[ { "created": "Thu, 26 Oct 2006 10:01:38 GMT", "version": "v1" } ]
2007-07-13
[ [ "Sahai", "Anant", "" ] ]
It is well known that orthogonal coding can be used to approach the Shannon capacity of the power-constrained AWGN channel without a bandwidth constraint. This correspondence describes a semi-orthogonal variation of pulse position modulation that is sequential in nature -- bits can be ``streamed across'' without having to buffer up blocks of bits at the transmitter. ML decoding results in an exponentially small probability of error as a function of tolerated receiver delay and thus eventually a zero probability of error on every transmitted bit. In the high-rate regime, a matching upper bound is given on the delay error exponent. We close with some comments on the case with feedback and the connections to the capacity per unit cost problem.
2103.09265
Harshitha Machiraju
Harshitha Machiraju, Oh-Hyeon Choung, Pascal Frossard, Michael. H Herzog
Bio-inspired Robustness: A Review
null
null
null
null
cs.CV cs.AI cs.LG
http://creativecommons.org/licenses/by-nc-nd/4.0/
Deep convolutional neural networks (DCNNs) have revolutionized computer vision and are often advocated as good models of the human visual system. However, there are currently many shortcomings of DCNNs, which preclude them as a model of human vision. For example, in the case of adversarial attacks, where adding small amounts of noise to an image, including an object, can lead to strong misclassification of that object. But for humans, the noise is often invisible. If vulnerability to adversarial noise cannot be fixed, DCNNs cannot be taken as serious models of human vision. Many studies have tried to add features of the human visual system to DCNNs to make them robust against adversarial attacks. However, it is not fully clear whether human vision inspired components increase robustness because performance evaluations of these novel components in DCNNs are often inconclusive. We propose a set of criteria for proper evaluation and analyze different models according to these criteria. We finally sketch future efforts to make DCCNs one step closer to the model of human vision.
[ { "created": "Tue, 16 Mar 2021 18:20:29 GMT", "version": "v1" } ]
2021-03-18
[ [ "Machiraju", "Harshitha", "" ], [ "Choung", "Oh-Hyeon", "" ], [ "Frossard", "Pascal", "" ], [ "Herzog", "Michael. H", "" ] ]
Deep convolutional neural networks (DCNNs) have revolutionized computer vision and are often advocated as good models of the human visual system. However, there are currently many shortcomings of DCNNs, which preclude them as a model of human vision. For example, in the case of adversarial attacks, where adding small amounts of noise to an image, including an object, can lead to strong misclassification of that object. But for humans, the noise is often invisible. If vulnerability to adversarial noise cannot be fixed, DCNNs cannot be taken as serious models of human vision. Many studies have tried to add features of the human visual system to DCNNs to make them robust against adversarial attacks. However, it is not fully clear whether human vision inspired components increase robustness because performance evaluations of these novel components in DCNNs are often inconclusive. We propose a set of criteria for proper evaluation and analyze different models according to these criteria. We finally sketch future efforts to make DCCNs one step closer to the model of human vision.
2304.13693
Thalia Santos De Santana
Thalia S. Santana, Taciana N. Kudo, Renato F. Bulc\~ao-Neto
Requirements Engineering, Software Testing and Education: A Systematic Mapping
20 pages, in Portuguese language
null
null
null
cs.SE
http://creativecommons.org/licenses/by-nc-nd/4.0/
The activities of requirements engineering and software testing are intrinsically related to each other, as these two areas are linked when seeking to specify and also ensure the expectations of a software product, with quality and on time. This systematic mapping study aims to verify how requirements and testing are being addressed together in the educational context.
[ { "created": "Wed, 26 Apr 2023 17:18:34 GMT", "version": "v1" } ]
2023-04-27
[ [ "Santana", "Thalia S.", "" ], [ "Kudo", "Taciana N.", "" ], [ "Bulcão-Neto", "Renato F.", "" ] ]
The activities of requirements engineering and software testing are intrinsically related to each other, as these two areas are linked when seeking to specify and also ensure the expectations of a software product, with quality and on time. This systematic mapping study aims to verify how requirements and testing are being addressed together in the educational context.
2003.01149
Piotr Franciszek Orzechowski
Piotr Franciszek Orzechowski, Christoph Burger and Martin Lauer
Decision-Making for Automated Vehicles Using a Hierarchical Behavior-Based Arbitration Scheme
null
null
10.1109/IV47402.2020.9304723
null
cs.RO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Behavior planning and decision-making are some of the biggest challenges for highly automated systems. A fully automated vehicle (AV) is confronted with numerous tactical and strategical choices. Most state-of-the-art AV platforms implement tactical and strategical behavior generation using finite state machines. However, these usually result in poor explainability, maintainability and scalability. Research in robotics has raised many architectures to mitigate these problems, most interestingly behavior-based systems and hybrid derivatives. Inspired by these approaches, we propose a hierarchical behavior-based architecture for tactical and strategical behavior generation in automated driving. It is a generalizing and scalable decision-making framework, utilizing modular behavior blocks to compose more complex behaviors in a bottom-up approach. The system is capable of combining a variety of scenario- and methodology-specific solutions, like POMDPs, RRT* or learning-based behavior, into one understandable and traceable architecture. We extend the hierarchical behavior-based arbitration concept to address scenarios where multiple behavior options are applicable but have no clear priority against each other. Then, we formulate the behavior generation stack for automated driving in urban and highway environments, incorporating parking and emergency behaviors as well. Finally, we illustrate our design in an explanatory evaluation.
[ { "created": "Mon, 2 Mar 2020 19:21:18 GMT", "version": "v1" }, { "created": "Mon, 11 May 2020 18:57:05 GMT", "version": "v2" }, { "created": "Tue, 1 Sep 2020 11:11:32 GMT", "version": "v3" }, { "created": "Fri, 5 Feb 2021 11:15:42 GMT", "version": "v4" } ]
2021-02-08
[ [ "Orzechowski", "Piotr Franciszek", "" ], [ "Burger", "Christoph", "" ], [ "Lauer", "Martin", "" ] ]
Behavior planning and decision-making are some of the biggest challenges for highly automated systems. A fully automated vehicle (AV) is confronted with numerous tactical and strategical choices. Most state-of-the-art AV platforms implement tactical and strategical behavior generation using finite state machines. However, these usually result in poor explainability, maintainability and scalability. Research in robotics has raised many architectures to mitigate these problems, most interestingly behavior-based systems and hybrid derivatives. Inspired by these approaches, we propose a hierarchical behavior-based architecture for tactical and strategical behavior generation in automated driving. It is a generalizing and scalable decision-making framework, utilizing modular behavior blocks to compose more complex behaviors in a bottom-up approach. The system is capable of combining a variety of scenario- and methodology-specific solutions, like POMDPs, RRT* or learning-based behavior, into one understandable and traceable architecture. We extend the hierarchical behavior-based arbitration concept to address scenarios where multiple behavior options are applicable but have no clear priority against each other. Then, we formulate the behavior generation stack for automated driving in urban and highway environments, incorporating parking and emergency behaviors as well. Finally, we illustrate our design in an explanatory evaluation.
2306.11767
Marvin Schieseck
Marvin Schieseck, Philip Topalis, Alexander Fay
A Graphical Modeling Language for Artificial Intelligence Applications in Automation Systems
null
null
null
null
cs.AI cs.SY eess.SY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Artificial Intelligence (AI) applications in automation systems are usually distributed systems whose development and integration involve several experts. Each expert uses its own domain-specific modeling language and tools to model the system elements. An interdisciplinary graphical modeling language that enables the modeling of an AI application as an overall system comprehensible to all disciplines does not yet exist. As a result, there is often a lack of interdisciplinary system understanding, leading to increased development, integration, and maintenance efforts. This paper therefore presents a graphical modeling language that enables consistent and understandable modeling of AI applications in automation systems at system level. This makes it possible to subdivide individual subareas into domain specific subsystems and thus reduce the existing efforts.
[ { "created": "Tue, 20 Jun 2023 12:06:41 GMT", "version": "v1" } ]
2023-06-22
[ [ "Schieseck", "Marvin", "" ], [ "Topalis", "Philip", "" ], [ "Fay", "Alexander", "" ] ]
Artificial Intelligence (AI) applications in automation systems are usually distributed systems whose development and integration involve several experts. Each expert uses its own domain-specific modeling language and tools to model the system elements. An interdisciplinary graphical modeling language that enables the modeling of an AI application as an overall system comprehensible to all disciplines does not yet exist. As a result, there is often a lack of interdisciplinary system understanding, leading to increased development, integration, and maintenance efforts. This paper therefore presents a graphical modeling language that enables consistent and understandable modeling of AI applications in automation systems at system level. This makes it possible to subdivide individual subareas into domain specific subsystems and thus reduce the existing efforts.
2211.02283
Jincheng Dai
Zixuan Xiao, Shengshi Yao, Jincheng Dai, Sixian Wang, Kai Niu, Ping Zhang
Wireless Deep Speech Semantic Transmission
null
null
null
null
cs.SD cs.IT eess.AS math.IT
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper, we propose a new class of high-efficiency semantic coded transmission methods for end-to-end speech transmission over wireless channels. We name the whole system as deep speech semantic transmission (DSST). Specifically, we introduce a nonlinear transform to map the speech source to semantic latent space and feed semantic features into source-channel encoder to generate the channel-input sequence. Guided by the variational modeling idea, we build an entropy model on the latent space to estimate the importance diversity among semantic feature embeddings. Accordingly, these semantic features of different importance can be allocated with different coding rates reasonably, which maximizes the system coding gain. Furthermore, we introduce a channel signal-to-noise ratio (SNR) adaptation mechanism such that a single model can be applied over various channel states. The end-to-end optimization of our model leads to a flexible rate-distortion (RD) trade-off, supporting versatile wireless speech semantic transmission. Experimental results verify that our DSST system clearly outperforms current engineered speech transmission systems on both objective and subjective metrics. Compared with existing neural speech semantic transmission methods, our model saves up to 75% of channel bandwidth costs when achieving the same quality. An intuitive comparison of audio demos can be found at https://ximoo123.github.io/DSST.
[ { "created": "Fri, 4 Nov 2022 06:49:42 GMT", "version": "v1" } ]
2022-11-07
[ [ "Xiao", "Zixuan", "" ], [ "Yao", "Shengshi", "" ], [ "Dai", "Jincheng", "" ], [ "Wang", "Sixian", "" ], [ "Niu", "Kai", "" ], [ "Zhang", "Ping", "" ] ]
In this paper, we propose a new class of high-efficiency semantic coded transmission methods for end-to-end speech transmission over wireless channels. We name the whole system as deep speech semantic transmission (DSST). Specifically, we introduce a nonlinear transform to map the speech source to semantic latent space and feed semantic features into source-channel encoder to generate the channel-input sequence. Guided by the variational modeling idea, we build an entropy model on the latent space to estimate the importance diversity among semantic feature embeddings. Accordingly, these semantic features of different importance can be allocated with different coding rates reasonably, which maximizes the system coding gain. Furthermore, we introduce a channel signal-to-noise ratio (SNR) adaptation mechanism such that a single model can be applied over various channel states. The end-to-end optimization of our model leads to a flexible rate-distortion (RD) trade-off, supporting versatile wireless speech semantic transmission. Experimental results verify that our DSST system clearly outperforms current engineered speech transmission systems on both objective and subjective metrics. Compared with existing neural speech semantic transmission methods, our model saves up to 75% of channel bandwidth costs when achieving the same quality. An intuitive comparison of audio demos can be found at https://ximoo123.github.io/DSST.
1609.00686
Makoto Naruse
Makoto Naruse, Martin Berthel, Aur\'elien Drezet, Serge Huant, Hirokazu Hori, Song-Ju Kim
Single photon in hierarchical architecture for physical reinforcement learning: Photon intelligence
null
null
null
null
cs.LG physics.optics quant-ph
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Understanding and using natural processes for intelligent functionalities, referred to as natural intelligence, has recently attracted interest from a variety of fields, including post-silicon computing for artificial intelligence and decision making in the behavioural sciences. In a past study, we successfully used the wave-particle duality of single photons to solve the two-armed bandit problem, which constitutes the foundation of reinforcement learning and decision making. In this study, we propose and confirm a hierarchical architecture for single-photon-based reinforcement learning and decision making that verifies the scalability of the principle. Specifically, the four-armed bandit problem is solved given zero prior knowledge in a two-layer hierarchical architecture, where polarization is autonomously adapted in order to effect adequate decision making using single-photon measurements. In the hierarchical structure, the notion of layer-dependent decisions emerges. The optimal solutions in the coarse layer and in the fine layer, however, conflict with each other in some contradictive problems. We show that while what we call a tournament strategy resolves such contradictions, the probabilistic nature of single photons allows for the direct location of the optimal solution even for contradictive problems, hence manifesting the exploration ability of single photons. This study provides insights into photon intelligence in hierarchical architectures for future artificial intelligence as well as the potential of natural processes for intelligent functionalities.
[ { "created": "Thu, 1 Sep 2016 09:32:29 GMT", "version": "v1" } ]
2016-09-05
[ [ "Naruse", "Makoto", "" ], [ "Berthel", "Martin", "" ], [ "Drezet", "Aurélien", "" ], [ "Huant", "Serge", "" ], [ "Hori", "Hirokazu", "" ], [ "Kim", "Song-Ju", "" ] ]
Understanding and using natural processes for intelligent functionalities, referred to as natural intelligence, has recently attracted interest from a variety of fields, including post-silicon computing for artificial intelligence and decision making in the behavioural sciences. In a past study, we successfully used the wave-particle duality of single photons to solve the two-armed bandit problem, which constitutes the foundation of reinforcement learning and decision making. In this study, we propose and confirm a hierarchical architecture for single-photon-based reinforcement learning and decision making that verifies the scalability of the principle. Specifically, the four-armed bandit problem is solved given zero prior knowledge in a two-layer hierarchical architecture, where polarization is autonomously adapted in order to effect adequate decision making using single-photon measurements. In the hierarchical structure, the notion of layer-dependent decisions emerges. The optimal solutions in the coarse layer and in the fine layer, however, conflict with each other in some contradictive problems. We show that while what we call a tournament strategy resolves such contradictions, the probabilistic nature of single photons allows for the direct location of the optimal solution even for contradictive problems, hence manifesting the exploration ability of single photons. This study provides insights into photon intelligence in hierarchical architectures for future artificial intelligence as well as the potential of natural processes for intelligent functionalities.
2107.02467
Hui Liu
J. Wang, X. Liu, S. Shen, L. Deng, H. Liu*
DeepDDS: deep graph neural network with attention mechanism to predict synergistic drug combinations
null
null
null
null
cs.LG q-bio.QM
http://creativecommons.org/licenses/by-nc-sa/4.0/
Drug combination therapy has become a increasingly promising method in the treatment of cancer. However, the number of possible drug combinations is so huge that it is hard to screen synergistic drug combinations through wet-lab experiments. Therefore, computational screening has become an important way to prioritize drug combinations. Graph neural network have recently shown remarkable performance in the prediction of compound-protein interactions, but it has not been applied to the screening of drug combinations. In this paper, we proposed a deep learning model based on graph neural networks and attention mechanism to identify drug combinations that can effectively inhibit the viability of specific cancer cells. The feature embeddings of drug molecule structure and gene expression profiles were taken as input to multi-layer feedforward neural network to identify the synergistic drug combinations. We compared DeepDDS with classical machine learning methods and other deep learning-based methods on benchmark data set, and the leave-one-out experimental results showed that DeepDDS achieved better performance than competitive methods. Also, on an independent test set released by well-known pharmaceutical enterprise AstraZeneca, DeepDDS was superior to competitive methods by more than 16\% predictive precision. Furthermore, we explored the interpretability of the graph attention network, and found the correlation matrix of atomic features revealed important chemical substructures of drugs. We believed that DeepDDS is an effective tool that prioritized synergistic drug combinations for further wet-lab experiment validation.
[ { "created": "Tue, 6 Jul 2021 08:25:43 GMT", "version": "v1" } ]
2021-07-07
[ [ "Wang", "J.", "" ], [ "Liu", "X.", "" ], [ "Shen", "S.", "" ], [ "Deng", "L.", "" ], [ "Liu*", "H.", "" ] ]
Drug combination therapy has become a increasingly promising method in the treatment of cancer. However, the number of possible drug combinations is so huge that it is hard to screen synergistic drug combinations through wet-lab experiments. Therefore, computational screening has become an important way to prioritize drug combinations. Graph neural network have recently shown remarkable performance in the prediction of compound-protein interactions, but it has not been applied to the screening of drug combinations. In this paper, we proposed a deep learning model based on graph neural networks and attention mechanism to identify drug combinations that can effectively inhibit the viability of specific cancer cells. The feature embeddings of drug molecule structure and gene expression profiles were taken as input to multi-layer feedforward neural network to identify the synergistic drug combinations. We compared DeepDDS with classical machine learning methods and other deep learning-based methods on benchmark data set, and the leave-one-out experimental results showed that DeepDDS achieved better performance than competitive methods. Also, on an independent test set released by well-known pharmaceutical enterprise AstraZeneca, DeepDDS was superior to competitive methods by more than 16\% predictive precision. Furthermore, we explored the interpretability of the graph attention network, and found the correlation matrix of atomic features revealed important chemical substructures of drugs. We believed that DeepDDS is an effective tool that prioritized synergistic drug combinations for further wet-lab experiment validation.
1603.06756
Wayes Tushar
Wayes Tushar, Chau Yuen, Bo Chai, Shisheng Huang, Kristin L. Wood, See Gim Kerk and Zaiyue Yang
Smart Grid Testbed for Demand Focused Energy Management in End User Environments
2016
null
null
null
cs.SY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Successful deployment of smart grids necessitates experimental validities of their state-of-the-art designs in two-way communications, real-time demand response and monitoring of consumers' energy usage behavior. The objective is to observe consumers' energy usage pattern and exploit this information to assist the grid in designing incentives, energy management mechanisms, and real-time demand response protocols; so as help the grid achieving lower costs and improve energy supply stability. Further, by feeding the observed information back to the consumers instantaneously, it is also possible to promote energy efficient behavior among the users. To this end, this paper performs a literature survey on smart grid testbeds around the world, and presents the main accomplishments towards realizing a smart grid testbed at the Singapore University of Technology and Design (SUTD). The testbed is able to monitor, analyze and evaluate smart grid communication network design and control mechanisms, and test the suitability of various communications networks for both residential and commercial buildings. The testbeds are deployed within the SUTD student dormitories and the main university campus to monitor and record end-user energy consumption in real-time, which will enable us to design incentives, control algorithms and real-time demand response schemes. The testbed also provides an effective channel to evaluate the needs on communication networks to support various smart grid applications. In addition, our initial results demonstrate that our testbed can provide an effective platform to identify energy wastage, and prompt the needs of a secure communications channel as the energy usage pattern can provide privacy related information on individual user.
[ { "created": "Tue, 22 Mar 2016 12:26:48 GMT", "version": "v1" } ]
2016-03-23
[ [ "Tushar", "Wayes", "" ], [ "Yuen", "Chau", "" ], [ "Chai", "Bo", "" ], [ "Huang", "Shisheng", "" ], [ "Wood", "Kristin L.", "" ], [ "Kerk", "See Gim", "" ], [ "Yang", "Zaiyue", "" ] ]
Successful deployment of smart grids necessitates experimental validities of their state-of-the-art designs in two-way communications, real-time demand response and monitoring of consumers' energy usage behavior. The objective is to observe consumers' energy usage pattern and exploit this information to assist the grid in designing incentives, energy management mechanisms, and real-time demand response protocols; so as help the grid achieving lower costs and improve energy supply stability. Further, by feeding the observed information back to the consumers instantaneously, it is also possible to promote energy efficient behavior among the users. To this end, this paper performs a literature survey on smart grid testbeds around the world, and presents the main accomplishments towards realizing a smart grid testbed at the Singapore University of Technology and Design (SUTD). The testbed is able to monitor, analyze and evaluate smart grid communication network design and control mechanisms, and test the suitability of various communications networks for both residential and commercial buildings. The testbeds are deployed within the SUTD student dormitories and the main university campus to monitor and record end-user energy consumption in real-time, which will enable us to design incentives, control algorithms and real-time demand response schemes. The testbed also provides an effective channel to evaluate the needs on communication networks to support various smart grid applications. In addition, our initial results demonstrate that our testbed can provide an effective platform to identify energy wastage, and prompt the needs of a secure communications channel as the energy usage pattern can provide privacy related information on individual user.
1506.00572
Scott A. Hale
Han-Teng Liao, King-wa Fu, Scott A. Hale
How much is said in a microblog? A multilingual inquiry based on Weibo and Twitter
9 pages, 4 figures WebSci 2015
null
10.1145/2786451.2786486
null
cs.SI cs.CL cs.CY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
This paper presents a multilingual study on, per single post of microblog text, (a) how much can be said, (b) how much is written in terms of characters and bytes, and (c) how much is said in terms of information content in posts by different organizations in different languages. Focusing on three different languages (English, Chinese, and Japanese), this research analyses Weibo and Twitter accounts of major embassies and news agencies. We first establish our criterion for quantifying "how much can be said" in a digital text based on the openly available Universal Declaration of Human Rights and the translated subtitles from TED talks. These parallel corpora allow us to determine the number of characters and bits needed to represent the same content in different languages and character encodings. We then derive the amount of information that is actually contained in microblog posts authored by selected accounts on Weibo and Twitter. Our results confirm that languages with larger character sets such as Chinese and Japanese contain more information per character than English, but the actual information content contained within a microblog text varies depending on both the type of organization and the language of the post. We conclude with a discussion on the design implications of microblog text limits for different languages.
[ { "created": "Mon, 1 Jun 2015 17:06:00 GMT", "version": "v1" }, { "created": "Sat, 13 Jun 2015 14:37:25 GMT", "version": "v2" } ]
2015-06-16
[ [ "Liao", "Han-Teng", "" ], [ "Fu", "King-wa", "" ], [ "Hale", "Scott A.", "" ] ]
This paper presents a multilingual study on, per single post of microblog text, (a) how much can be said, (b) how much is written in terms of characters and bytes, and (c) how much is said in terms of information content in posts by different organizations in different languages. Focusing on three different languages (English, Chinese, and Japanese), this research analyses Weibo and Twitter accounts of major embassies and news agencies. We first establish our criterion for quantifying "how much can be said" in a digital text based on the openly available Universal Declaration of Human Rights and the translated subtitles from TED talks. These parallel corpora allow us to determine the number of characters and bits needed to represent the same content in different languages and character encodings. We then derive the amount of information that is actually contained in microblog posts authored by selected accounts on Weibo and Twitter. Our results confirm that languages with larger character sets such as Chinese and Japanese contain more information per character than English, but the actual information content contained within a microblog text varies depending on both the type of organization and the language of the post. We conclude with a discussion on the design implications of microblog text limits for different languages.
2306.07282
Karsten Roth
Karsten Roth, Jae Myung Kim, A. Sophia Koepke, Oriol Vinyals, Cordelia Schmid, Zeynep Akata
Waffling around for Performance: Visual Classification with Random Words and Broad Concepts
Accepted to ICCV 2023. Main paper with 9 pages
null
null
null
cs.CV cs.LG
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The visual classification performance of vision-language models such as CLIP has been shown to benefit from additional semantic knowledge from large language models (LLMs) such as GPT-3. In particular, averaging over LLM-generated class descriptors, e.g. "waffle, which has a round shape", can notably improve generalization performance. In this work, we critically study this behavior and propose WaffleCLIP, a framework for zero-shot visual classification which simply replaces LLM-generated descriptors with random character and word descriptors. Without querying external models, we achieve comparable performance gains on a large number of visual classification tasks. This allows WaffleCLIP to both serve as a low-cost alternative, as well as a sanity check for any future LLM-based vision-language model extensions. We conduct an extensive experimental study on the impact and shortcomings of additional semantics introduced with LLM-generated descriptors, and showcase how - if available - semantic context is better leveraged by querying LLMs for high-level concepts, which we show can be done to jointly resolve potential class name ambiguities. Code is available here: https://github.com/ExplainableML/WaffleCLIP.
[ { "created": "Mon, 12 Jun 2023 17:59:48 GMT", "version": "v1" }, { "created": "Thu, 17 Aug 2023 02:27:32 GMT", "version": "v2" } ]
2023-08-21
[ [ "Roth", "Karsten", "" ], [ "Kim", "Jae Myung", "" ], [ "Koepke", "A. Sophia", "" ], [ "Vinyals", "Oriol", "" ], [ "Schmid", "Cordelia", "" ], [ "Akata", "Zeynep", "" ] ]
The visual classification performance of vision-language models such as CLIP has been shown to benefit from additional semantic knowledge from large language models (LLMs) such as GPT-3. In particular, averaging over LLM-generated class descriptors, e.g. "waffle, which has a round shape", can notably improve generalization performance. In this work, we critically study this behavior and propose WaffleCLIP, a framework for zero-shot visual classification which simply replaces LLM-generated descriptors with random character and word descriptors. Without querying external models, we achieve comparable performance gains on a large number of visual classification tasks. This allows WaffleCLIP to both serve as a low-cost alternative, as well as a sanity check for any future LLM-based vision-language model extensions. We conduct an extensive experimental study on the impact and shortcomings of additional semantics introduced with LLM-generated descriptors, and showcase how - if available - semantic context is better leveraged by querying LLMs for high-level concepts, which we show can be done to jointly resolve potential class name ambiguities. Code is available here: https://github.com/ExplainableML/WaffleCLIP.
2203.07463
Ramin Raziperchikolaei
Ramin Raziperchikolaei and Young-joo Chung
Simultaneous Learning of the Inputs and Parameters in Neural Collaborative Filtering
null
null
null
null
cs.IR cs.LG
http://creativecommons.org/licenses/by/4.0/
Neural network-based collaborative filtering systems focus on designing network architectures to learn better representations while fixing the input to the user/item interaction vectors and/or ID. In this paper, we first show that the non-zero elements of the inputs are learnable parameters that determine the weights in combining the user/item embeddings, and fixing them limits the power of the models in learning the representations. Then, we propose to learn the value of the non-zero elements of the inputs jointly with the neural network parameters. We analyze the model complexity and the empirical risk of our approach and prove that learning the input leads to a better generalization bound. Our experiments on several real-world datasets show that our method outperforms the state-of-the-art methods, even using shallow network structures with a smaller number of layers and parameters.
[ { "created": "Mon, 14 Mar 2022 19:47:38 GMT", "version": "v1" } ]
2022-03-16
[ [ "Raziperchikolaei", "Ramin", "" ], [ "Chung", "Young-joo", "" ] ]
Neural network-based collaborative filtering systems focus on designing network architectures to learn better representations while fixing the input to the user/item interaction vectors and/or ID. In this paper, we first show that the non-zero elements of the inputs are learnable parameters that determine the weights in combining the user/item embeddings, and fixing them limits the power of the models in learning the representations. Then, we propose to learn the value of the non-zero elements of the inputs jointly with the neural network parameters. We analyze the model complexity and the empirical risk of our approach and prove that learning the input leads to a better generalization bound. Our experiments on several real-world datasets show that our method outperforms the state-of-the-art methods, even using shallow network structures with a smaller number of layers and parameters.
1911.09548
Martin Schwalsberger
Martin Neum\"uller, Martin Schwalsberger
A parallel space-time multigrid method for the eddy-current equation
null
null
null
null
cs.CE cs.NA math.NA
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We expand the applicabilities and capabilities of an already existing space-time parallel method based on a block Jacobi smoother. First we formulate a more detailed criterion for spatial coarsening, which enables the method to deal with unstructured meshes and varying material parameters. Further we investigate the application to the eddy-current equation, where the non-trivial kernel of the curl operator causes severe problems. This is remedied with a new nodal auxiliary space correction. We proceed to identify convergence rates by local Fourier analysis and numerical experiments. Finally, we present a numerical experiment which demonstrates its excellent scaling properties.
[ { "created": "Thu, 21 Nov 2019 15:43:40 GMT", "version": "v1" } ]
2019-11-22
[ [ "Neumüller", "Martin", "" ], [ "Schwalsberger", "Martin", "" ] ]
We expand the applicabilities and capabilities of an already existing space-time parallel method based on a block Jacobi smoother. First we formulate a more detailed criterion for spatial coarsening, which enables the method to deal with unstructured meshes and varying material parameters. Further we investigate the application to the eddy-current equation, where the non-trivial kernel of the curl operator causes severe problems. This is remedied with a new nodal auxiliary space correction. We proceed to identify convergence rates by local Fourier analysis and numerical experiments. Finally, we present a numerical experiment which demonstrates its excellent scaling properties.
2208.13137
Priyabarta Karmakar PhD
Priyabrata Karmakar, Manzur Murshed, Manoranjan Paul, David Taubman
Efficient Motion Modelling with Variable-sized blocks from Hierarchical Cuboidal Partitioning
null
null
null
null
cs.CV cs.MM
http://creativecommons.org/licenses/by/4.0/
Motion modelling with block-based architecture has been widely used in video coding where a frame is divided into fixed-sized blocks that are motion compensated independently. This often leads to coding inefficiency as fixed-sized blocks hardly align with the object boundaries. Although hierarchical block-partitioning has been introduced to address this, the increased number of motion vectors limits the benefit. Recently, approximate segmentation of images with cuboidal partitioning has gained popularity. Not only are the variable-sized rectangular segments (cuboids) readily amenable to block-based image/video coding techniques, but they are also capable of aligning well with the object boundaries. This is because cuboidal partitioning is based on a homogeneity constraint, minimising the sum of squared errors (SSE). In this paper, we have investigated the potential of cuboids in motion modelling against the fixed-sized blocks used in scalable video coding. Specifically, we have constructed motion-compensated current frame using the cuboidal partitioning information of the anchor frame in a group-of-picture (GOP). The predicted current frame has then been used as the base layer while encoding the current frame as an enhancement layer using the scalable HEVC encoder. Experimental results confirm 6.71%-10.90% bitrate savings on 4K video sequences.
[ { "created": "Sun, 28 Aug 2022 04:13:58 GMT", "version": "v1" } ]
2022-08-30
[ [ "Karmakar", "Priyabrata", "" ], [ "Murshed", "Manzur", "" ], [ "Paul", "Manoranjan", "" ], [ "Taubman", "David", "" ] ]
Motion modelling with block-based architecture has been widely used in video coding where a frame is divided into fixed-sized blocks that are motion compensated independently. This often leads to coding inefficiency as fixed-sized blocks hardly align with the object boundaries. Although hierarchical block-partitioning has been introduced to address this, the increased number of motion vectors limits the benefit. Recently, approximate segmentation of images with cuboidal partitioning has gained popularity. Not only are the variable-sized rectangular segments (cuboids) readily amenable to block-based image/video coding techniques, but they are also capable of aligning well with the object boundaries. This is because cuboidal partitioning is based on a homogeneity constraint, minimising the sum of squared errors (SSE). In this paper, we have investigated the potential of cuboids in motion modelling against the fixed-sized blocks used in scalable video coding. Specifically, we have constructed motion-compensated current frame using the cuboidal partitioning information of the anchor frame in a group-of-picture (GOP). The predicted current frame has then been used as the base layer while encoding the current frame as an enhancement layer using the scalable HEVC encoder. Experimental results confirm 6.71%-10.90% bitrate savings on 4K video sequences.
1007.0918
Jonathan Heusser
Jonathan Heusser and Pasquale Malacaria
Quantifying Information Leak Vulnerabilities
submitted, under review
null
null
null
cs.CR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Leakage of confidential information represents a serious security risk. Despite a number of novel, theoretical advances, it has been unclear if and how quantitative approaches to measuring leakage of confidential information could be applied to substantial, real-world programs. This is mostly due to the high complexity of computing precise leakage quantities. In this paper, we introduce a technique which makes it possible to decide if a program conforms to a quantitative policy which scales to large state-spaces with the help of bounded model checking. Our technique is applied to a number of officially reported information leak vulnerabilities in the Linux Kernel. Additionally, we also analysed authentication routines in the Secure Remote Password suite and of a Internet Message Support Protocol implementation. Our technique shows when there is unacceptable leakage; the same technique is also used to verify, for the first time, that the applied software patches indeed plug the information leaks. This is the first demonstration of quantitative information flow addressing security concerns of real-world industrial programs.
[ { "created": "Tue, 6 Jul 2010 15:12:12 GMT", "version": "v1" } ]
2010-07-07
[ [ "Heusser", "Jonathan", "" ], [ "Malacaria", "Pasquale", "" ] ]
Leakage of confidential information represents a serious security risk. Despite a number of novel, theoretical advances, it has been unclear if and how quantitative approaches to measuring leakage of confidential information could be applied to substantial, real-world programs. This is mostly due to the high complexity of computing precise leakage quantities. In this paper, we introduce a technique which makes it possible to decide if a program conforms to a quantitative policy which scales to large state-spaces with the help of bounded model checking. Our technique is applied to a number of officially reported information leak vulnerabilities in the Linux Kernel. Additionally, we also analysed authentication routines in the Secure Remote Password suite and of a Internet Message Support Protocol implementation. Our technique shows when there is unacceptable leakage; the same technique is also used to verify, for the first time, that the applied software patches indeed plug the information leaks. This is the first demonstration of quantitative information flow addressing security concerns of real-world industrial programs.
1910.05798
Jefferson Silva
Jefferson O. Silva, Igor Wiese, Daniel M. German, Christoph Treude, Marco A. Gerosa, Igor Steinmacher
Google Summer of Code: Student Motivations and Contributions
30 pages
Journal of Systems and Software (JSS), V. 162, April 2020, 110487
10.1016/j.jss.2019.110487
null
cs.SE
http://creativecommons.org/licenses/by/4.0/
Several open source software (OSS) projects expect to foster newcomers' onboarding and to receive contributions by participating in engagement programs, like Summers of Code. However, there is little empirical evidence showing why students join such programs. In this paper, we study the well-established Google Summer of Code (GSoC), which is a 3-month OSS engagement program that offers stipends and mentors to students willing to contribute to OSS projects. We combined a survey (students and mentors) and interviews (students) to understand what motivates students to enter GSoC. Our results show that students enter GSoC for an enriching experience, not necessarily to become frequent contributors. Our data suggest that, while the stipends are an important motivator, the students participate for work experience and the ability to attach the name of the supporting organization to their resum\'es. We also discuss practical implications for students, mentors, OSS projects, and Summer of Code programs.
[ { "created": "Sun, 13 Oct 2019 18:07:24 GMT", "version": "v1" } ]
2024-01-24
[ [ "Silva", "Jefferson O.", "" ], [ "Wiese", "Igor", "" ], [ "German", "Daniel M.", "" ], [ "Treude", "Christoph", "" ], [ "Gerosa", "Marco A.", "" ], [ "Steinmacher", "Igor", "" ] ]
Several open source software (OSS) projects expect to foster newcomers' onboarding and to receive contributions by participating in engagement programs, like Summers of Code. However, there is little empirical evidence showing why students join such programs. In this paper, we study the well-established Google Summer of Code (GSoC), which is a 3-month OSS engagement program that offers stipends and mentors to students willing to contribute to OSS projects. We combined a survey (students and mentors) and interviews (students) to understand what motivates students to enter GSoC. Our results show that students enter GSoC for an enriching experience, not necessarily to become frequent contributors. Our data suggest that, while the stipends are an important motivator, the students participate for work experience and the ability to attach the name of the supporting organization to their resum\'es. We also discuss practical implications for students, mentors, OSS projects, and Summer of Code programs.
2210.10040
Nikil Selvam
Nikil Roashan Selvam, Sunipa Dev, Daniel Khashabi, Tushar Khot, Kai-Wei Chang
The Tail Wagging the Dog: Dataset Construction Biases of Social Bias Benchmarks
ACL 2023
null
null
null
cs.CL cs.CY cs.LG cs.SI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
How reliably can we trust the scores obtained from social bias benchmarks as faithful indicators of problematic social biases in a given language model? In this work, we study this question by contrasting social biases with non-social biases stemming from choices made during dataset construction that might not even be discernible to the human eye. To do so, we empirically simulate various alternative constructions for a given benchmark based on innocuous modifications (such as paraphrasing or random-sampling) that maintain the essence of their social bias. On two well-known social bias benchmarks (Winogender and BiasNLI) we observe that these shallow modifications have a surprising effect on the resulting degree of bias across various models. We hope these troubling observations motivate more robust measures of social biases.
[ { "created": "Tue, 18 Oct 2022 17:58:39 GMT", "version": "v1" }, { "created": "Fri, 16 Jun 2023 18:35:13 GMT", "version": "v2" } ]
2023-06-21
[ [ "Selvam", "Nikil Roashan", "" ], [ "Dev", "Sunipa", "" ], [ "Khashabi", "Daniel", "" ], [ "Khot", "Tushar", "" ], [ "Chang", "Kai-Wei", "" ] ]
How reliably can we trust the scores obtained from social bias benchmarks as faithful indicators of problematic social biases in a given language model? In this work, we study this question by contrasting social biases with non-social biases stemming from choices made during dataset construction that might not even be discernible to the human eye. To do so, we empirically simulate various alternative constructions for a given benchmark based on innocuous modifications (such as paraphrasing or random-sampling) that maintain the essence of their social bias. On two well-known social bias benchmarks (Winogender and BiasNLI) we observe that these shallow modifications have a surprising effect on the resulting degree of bias across various models. We hope these troubling observations motivate more robust measures of social biases.
2110.10246
Nicole Forsgren
Nicole Forsgren, Bas Alberts, Kevin Backhouse, Grey Baker, Greg Cecarelli, Derek Jedamski, Scot Kelly, Clair Sullivan
2020 State of the Octoverse: Securing the World's Software
published by GitHub
null
null
null
cs.SE
http://creativecommons.org/licenses/by/4.0/
Open source is the connective tissue for much of the information economy. You would be hard-pressed to find a scenario where your data does not pass through at least one open source component. Many of the services and technology we all rely on, from banking to healthcare, also rely on open source software. The artifacts of open source code serve as critical i infrastructure for much of the global economy, making the security of open source software mission-critical to the world.
[ { "created": "Tue, 19 Oct 2021 20:32:32 GMT", "version": "v1" } ]
2021-10-22
[ [ "Forsgren", "Nicole", "" ], [ "Alberts", "Bas", "" ], [ "Backhouse", "Kevin", "" ], [ "Baker", "Grey", "" ], [ "Cecarelli", "Greg", "" ], [ "Jedamski", "Derek", "" ], [ "Kelly", "Scot", "" ], [ "Sullivan", "Clair", "" ] ]
Open source is the connective tissue for much of the information economy. You would be hard-pressed to find a scenario where your data does not pass through at least one open source component. Many of the services and technology we all rely on, from banking to healthcare, also rely on open source software. The artifacts of open source code serve as critical i infrastructure for much of the global economy, making the security of open source software mission-critical to the world.
2107.12562
Shifeng Pan
Shifeng Pan and Lei He
Cross-speaker Style Transfer with Prosody Bottleneck in Neural Speech Synthesis
in Proceedings of INTERSPEECH 2021
null
null
null
cs.SD cs.LG eess.AS
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Cross-speaker style transfer is crucial to the applications of multi-style and expressive speech synthesis at scale. It does not require the target speakers to be experts in expressing all styles and to collect corresponding recordings for model training. However, the performances of existing style transfer methods are still far behind real application needs. The root causes are mainly twofold. Firstly, the style embedding extracted from single reference speech can hardly provide fine-grained and appropriate prosody information for arbitrary text to synthesize. Secondly, in these models the content/text, prosody, and speaker timbre are usually highly entangled, it's therefore not realistic to expect a satisfied result when freely combining these components, such as to transfer speaking style between speakers. In this paper, we propose a cross-speaker style transfer text-to-speech (TTS) model with explicit prosody bottleneck. The prosody bottleneck builds up the kernels accounting for speaking style robustly, and disentangles the prosody from content and speaker timbre, therefore guarantees high quality cross-speaker style transfer. Evaluation result shows the proposed method even achieves on-par performance with source speaker's speaker-dependent (SD) model in objective measurement of prosody, and significantly outperforms the cycle consistency and GMVAE-based baselines in objective and subjective evaluations.
[ { "created": "Tue, 27 Jul 2021 02:43:57 GMT", "version": "v1" } ]
2021-07-28
[ [ "Pan", "Shifeng", "" ], [ "He", "Lei", "" ] ]
Cross-speaker style transfer is crucial to the applications of multi-style and expressive speech synthesis at scale. It does not require the target speakers to be experts in expressing all styles and to collect corresponding recordings for model training. However, the performances of existing style transfer methods are still far behind real application needs. The root causes are mainly twofold. Firstly, the style embedding extracted from single reference speech can hardly provide fine-grained and appropriate prosody information for arbitrary text to synthesize. Secondly, in these models the content/text, prosody, and speaker timbre are usually highly entangled, it's therefore not realistic to expect a satisfied result when freely combining these components, such as to transfer speaking style between speakers. In this paper, we propose a cross-speaker style transfer text-to-speech (TTS) model with explicit prosody bottleneck. The prosody bottleneck builds up the kernels accounting for speaking style robustly, and disentangles the prosody from content and speaker timbre, therefore guarantees high quality cross-speaker style transfer. Evaluation result shows the proposed method even achieves on-par performance with source speaker's speaker-dependent (SD) model in objective measurement of prosody, and significantly outperforms the cycle consistency and GMVAE-based baselines in objective and subjective evaluations.
1204.2989
Vijay Ganesh
Vijay Ganesh
STP/HAMPI and Computer Security
null
null
null
null
cs.CR cs.LO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In the past several years I have written two SMT solvers called STP and HAMPI that have found widespread use in computer security research by leading groups in academia, industry and the government. In this brief note I summarize the features of STP/HAMPI that make them particularly suited for computer security research, and a listing of some of the more important projects that use them.
[ { "created": "Thu, 12 Apr 2012 16:46:21 GMT", "version": "v1" } ]
2012-04-16
[ [ "Ganesh", "Vijay", "" ] ]
In the past several years I have written two SMT solvers called STP and HAMPI that have found widespread use in computer security research by leading groups in academia, industry and the government. In this brief note I summarize the features of STP/HAMPI that make them particularly suited for computer security research, and a listing of some of the more important projects that use them.
2303.03583
Siqi Fan
Siqi Fan, Zhe Wang, Xiaoliang Huo, Yan Wang, Jingjing Liu
Calibration-free BEV Representation for Infrastructure Perception
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Effective BEV object detection on infrastructure can greatly improve traffic scenes understanding and vehicle-toinfrastructure (V2I) cooperative perception. However, cameras installed on infrastructure have various postures, and previous BEV detection methods rely on accurate calibration, which is difficult for practical applications due to inevitable natural factors (e.g., wind and snow). In this paper, we propose a Calibration-free BEV Representation (CBR) network, which achieves 3D detection based on BEV representation without calibration parameters and additional depth supervision. Specifically, we utilize two multi-layer perceptrons for decoupling the features from perspective view to front view and birdeye view under boxes-induced foreground supervision. Then, a cross-view feature fusion module matches features from orthogonal views according to similarity and conducts BEV feature enhancement with front view features. Experimental results on DAIR-V2X demonstrate that CBR achieves acceptable performance without any camera parameters and is naturally not affected by calibration noises. We hope CBR can serve as a baseline for future research addressing practical challenges of infrastructure perception.
[ { "created": "Tue, 7 Mar 2023 01:31:06 GMT", "version": "v1" }, { "created": "Fri, 14 Apr 2023 02:45:05 GMT", "version": "v2" } ]
2023-04-17
[ [ "Fan", "Siqi", "" ], [ "Wang", "Zhe", "" ], [ "Huo", "Xiaoliang", "" ], [ "Wang", "Yan", "" ], [ "Liu", "Jingjing", "" ] ]
Effective BEV object detection on infrastructure can greatly improve traffic scenes understanding and vehicle-toinfrastructure (V2I) cooperative perception. However, cameras installed on infrastructure have various postures, and previous BEV detection methods rely on accurate calibration, which is difficult for practical applications due to inevitable natural factors (e.g., wind and snow). In this paper, we propose a Calibration-free BEV Representation (CBR) network, which achieves 3D detection based on BEV representation without calibration parameters and additional depth supervision. Specifically, we utilize two multi-layer perceptrons for decoupling the features from perspective view to front view and birdeye view under boxes-induced foreground supervision. Then, a cross-view feature fusion module matches features from orthogonal views according to similarity and conducts BEV feature enhancement with front view features. Experimental results on DAIR-V2X demonstrate that CBR achieves acceptable performance without any camera parameters and is naturally not affected by calibration noises. We hope CBR can serve as a baseline for future research addressing practical challenges of infrastructure perception.
cs/0205080
Matus Marko
M. Marko, M.A. Porter, A. Probst, C. Gershenson, A. Das
Transforming the World Wide Web into a Complexity-Based Semantic Network
6 pages, a manuscript for the ICCS 2002
null
null
null
cs.NI cs.IR
null
The aim of this paper is to introduce the idea of the Semantic Web to the Complexity community and set a basic ground for a project resulting in creation of Internet-based semantic network of Complexity-related information providers. Implementation of the Semantic Web technology would be of mutual benefit to both the participants and users and will confirm self-referencing power of the community to apply the products of its own research to itself. We first explain the logic of the transition and discuss important notions associated with the Semantic Web technology. We then present a brief outline of the project milestones.
[ { "created": "Fri, 31 May 2002 18:44:36 GMT", "version": "v1" }, { "created": "Sat, 1 Jun 2002 08:03:07 GMT", "version": "v2" } ]
2007-05-23
[ [ "Marko", "M.", "" ], [ "Porter", "M. A.", "" ], [ "Probst", "A.", "" ], [ "Gershenson", "C.", "" ], [ "Das", "A.", "" ] ]
The aim of this paper is to introduce the idea of the Semantic Web to the Complexity community and set a basic ground for a project resulting in creation of Internet-based semantic network of Complexity-related information providers. Implementation of the Semantic Web technology would be of mutual benefit to both the participants and users and will confirm self-referencing power of the community to apply the products of its own research to itself. We first explain the logic of the transition and discuss important notions associated with the Semantic Web technology. We then present a brief outline of the project milestones.
1711.03396
Chihao Zhang
Heng Guo, Chao Liao, Pinyan Lu, Chihao Zhang
Counting hypergraph colorings in the local lemma regime
v3: Constants Changed. Accepted to SICOMP
null
null
null
cs.DS cs.DM math.CO
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We give a fully polynomial-time approximation scheme (FPTAS) to count the number of $q$-colorings for $k$-uniform hypergraphs with maximum degree $\Delta$ if $k\ge 28$ and $q >357\Delta^{\frac{14}{k-14}}$ . We also obtain a polynomial-time almost uniform sampler if $q>931\Delta^{\frac{16}{k-16/3}}$. These are the first approximate counting and sampling algorithms in the regime $q\ll\Delta$ (for large $\Delta$ and $k$) without any additional assumptions. Our method is based on the recent work of Moitra (STOC, 2017). One important contribution of ours is to remove the dependency of $k$ and $\Delta$ in Moitra's approach.
[ { "created": "Thu, 9 Nov 2017 14:49:59 GMT", "version": "v1" }, { "created": "Sun, 12 Nov 2017 08:52:45 GMT", "version": "v2" }, { "created": "Fri, 31 May 2019 13:48:14 GMT", "version": "v3" } ]
2019-06-03
[ [ "Guo", "Heng", "" ], [ "Liao", "Chao", "" ], [ "Lu", "Pinyan", "" ], [ "Zhang", "Chihao", "" ] ]
We give a fully polynomial-time approximation scheme (FPTAS) to count the number of $q$-colorings for $k$-uniform hypergraphs with maximum degree $\Delta$ if $k\ge 28$ and $q >357\Delta^{\frac{14}{k-14}}$ . We also obtain a polynomial-time almost uniform sampler if $q>931\Delta^{\frac{16}{k-16/3}}$. These are the first approximate counting and sampling algorithms in the regime $q\ll\Delta$ (for large $\Delta$ and $k$) without any additional assumptions. Our method is based on the recent work of Moitra (STOC, 2017). One important contribution of ours is to remove the dependency of $k$ and $\Delta$ in Moitra's approach.
2308.06197
Angus Maiden MAAI(Prof)
Angus Maiden (1), Bahareh Nakisa (1) ((1) Deakin University)
Complex Facial Expression Recognition Using Deep Knowledge Distillation of Basic Features
13 pages, 9 figures, 6 tables, 3 algorithms. Code available at https://github.com/AngusMaiden/complex-FER
null
null
null
cs.CV cs.AI cs.LG
http://creativecommons.org/licenses/by/4.0/
Complex emotion recognition is a cognitive task that has so far eluded the same excellent performance of other tasks that are at or above the level of human cognition. Emotion recognition through facial expressions is particularly difficult due to the complexity of emotions expressed by the human face. For a machine to approach the same level of performance in complex facial expression recognition as a human, it may need to synthesise knowledge and understand new concepts in real-time, as humans do. Humans are able to learn new concepts using only few examples by distilling important information from memories. Inspired by human cognition and learning, we propose a novel continual learning method for complex facial expression recognition that can accurately recognise new compound expression classes using few training samples, by building on and retaining its knowledge of basic expression classes. In this work, we also use GradCAM visualisations to demonstrate the relationship between basic and compound facial expressions. Our method leverages this relationship through knowledge distillation and a novel Predictive Sorting Memory Replay, to achieve the current state-of-the-art in continual learning for complex facial expression recognition, with 74.28% Overall Accuracy on new classes. We also demonstrate that using continual learning for complex facial expression recognition achieves far better performance than non-continual learning methods, improving on state-of-the-art non-continual learning methods by 13.95%. Our work is also the first to apply few-shot learning to complex facial expression recognition, achieving the state-of-the-art with 100% accuracy using only a single training sample per class.
[ { "created": "Fri, 11 Aug 2023 15:42:48 GMT", "version": "v1" }, { "created": "Sun, 5 Nov 2023 23:34:25 GMT", "version": "v2" } ]
2023-11-07
[ [ "Maiden", "Angus", "", "Deakin University" ], [ "Nakisa", "Bahareh", "", "Deakin University" ] ]
Complex emotion recognition is a cognitive task that has so far eluded the same excellent performance of other tasks that are at or above the level of human cognition. Emotion recognition through facial expressions is particularly difficult due to the complexity of emotions expressed by the human face. For a machine to approach the same level of performance in complex facial expression recognition as a human, it may need to synthesise knowledge and understand new concepts in real-time, as humans do. Humans are able to learn new concepts using only few examples by distilling important information from memories. Inspired by human cognition and learning, we propose a novel continual learning method for complex facial expression recognition that can accurately recognise new compound expression classes using few training samples, by building on and retaining its knowledge of basic expression classes. In this work, we also use GradCAM visualisations to demonstrate the relationship between basic and compound facial expressions. Our method leverages this relationship through knowledge distillation and a novel Predictive Sorting Memory Replay, to achieve the current state-of-the-art in continual learning for complex facial expression recognition, with 74.28% Overall Accuracy on new classes. We also demonstrate that using continual learning for complex facial expression recognition achieves far better performance than non-continual learning methods, improving on state-of-the-art non-continual learning methods by 13.95%. Our work is also the first to apply few-shot learning to complex facial expression recognition, achieving the state-of-the-art with 100% accuracy using only a single training sample per class.
2402.01156
Yongkun Liu
Yongkun Liu, Jiachi Chen, Tingting Bi, John Grundy, Yanlin Wang, Jianxing Yu, Ting Chen, Yutian Tang, Zibin Zheng
An Empirical Study on Low Code Programming using Traditional vs Large Language Model Support
null
null
null
null
cs.SE
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Low-code programming (LCP) refers to programming using models at higher levels of abstraction, resulting in less manual and more efficient programming, and reduced learning effort for amateur developers. Many LCP tools have rapidly evolved and have benefited from the concepts of visual programming languages (VPLs) and programming by demonstration (PBD). With huge increase in interest in using large language models (LLMs) in software engineering, LLM-based LCP has began to become increasingly important. However, the technical principles and application scenarios of traditional approaches to LCP and LLM-based LCP are significantly different. Understanding these key differences and characteristics in the application of the two approaches to LCP by users is crucial for LCP providers in improving existing and developing new LCP tools, and in better assisting users in choosing the appropriate LCP technology. We conducted an empirical study of both traditional LCP and LLM-based LCP. We analyzed developers' discussions on Stack Overflow (SO) over the past three years and then explored the similarities and differences between traditional LCP and LLM-based LCP features and developer feedback. Our findings reveal that while traditional LCP and LLM-based LCP share common primary usage scenarios, they significantly differ in scope, limitations and usage throughout the software development lifecycle, particularly during the implementation phase. We also examine how LLMs impact and integrate with LCP, discussing the latest technological developments in LLM-based LCP, such as its integration with VPLs and the application of LLM Agents in software engineering.
[ { "created": "Fri, 2 Feb 2024 05:52:32 GMT", "version": "v1" }, { "created": "Thu, 6 Jun 2024 12:07:38 GMT", "version": "v2" } ]
2024-06-07
[ [ "Liu", "Yongkun", "" ], [ "Chen", "Jiachi", "" ], [ "Bi", "Tingting", "" ], [ "Grundy", "John", "" ], [ "Wang", "Yanlin", "" ], [ "Yu", "Jianxing", "" ], [ "Chen", "Ting", "" ], [ "Tang", "Yutian", "" ], [ "Zheng", "Zibin", "" ] ]
Low-code programming (LCP) refers to programming using models at higher levels of abstraction, resulting in less manual and more efficient programming, and reduced learning effort for amateur developers. Many LCP tools have rapidly evolved and have benefited from the concepts of visual programming languages (VPLs) and programming by demonstration (PBD). With huge increase in interest in using large language models (LLMs) in software engineering, LLM-based LCP has began to become increasingly important. However, the technical principles and application scenarios of traditional approaches to LCP and LLM-based LCP are significantly different. Understanding these key differences and characteristics in the application of the two approaches to LCP by users is crucial for LCP providers in improving existing and developing new LCP tools, and in better assisting users in choosing the appropriate LCP technology. We conducted an empirical study of both traditional LCP and LLM-based LCP. We analyzed developers' discussions on Stack Overflow (SO) over the past three years and then explored the similarities and differences between traditional LCP and LLM-based LCP features and developer feedback. Our findings reveal that while traditional LCP and LLM-based LCP share common primary usage scenarios, they significantly differ in scope, limitations and usage throughout the software development lifecycle, particularly during the implementation phase. We also examine how LLMs impact and integrate with LCP, discussing the latest technological developments in LLM-based LCP, such as its integration with VPLs and the application of LLM Agents in software engineering.
2406.14155
Dominik Stammbach
Dominik Stammbach and Philine Widmer and Eunjung Cho and Caglar Gulcehre and Elliott Ash
Aligning Large Language Models with Diverse Political Viewpoints
null
null
null
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
Large language models such as ChatGPT often exhibit striking political biases. If users query them about political information, they might take a normative stance and reinforce such biases. To overcome this, we align LLMs with diverse political viewpoints from 100,000 comments written by candidates running for national parliament in Switzerland. Such aligned models are able to generate more accurate political viewpoints from Swiss parties compared to commercial models such as ChatGPT. We also propose a procedure to generate balanced overviews from multiple viewpoints using such models.
[ { "created": "Thu, 20 Jun 2024 09:53:23 GMT", "version": "v1" } ]
2024-06-21
[ [ "Stammbach", "Dominik", "" ], [ "Widmer", "Philine", "" ], [ "Cho", "Eunjung", "" ], [ "Gulcehre", "Caglar", "" ], [ "Ash", "Elliott", "" ] ]
Large language models such as ChatGPT often exhibit striking political biases. If users query them about political information, they might take a normative stance and reinforce such biases. To overcome this, we align LLMs with diverse political viewpoints from 100,000 comments written by candidates running for national parliament in Switzerland. Such aligned models are able to generate more accurate political viewpoints from Swiss parties compared to commercial models such as ChatGPT. We also propose a procedure to generate balanced overviews from multiple viewpoints using such models.
2101.07768
Ehsan Zabardast
Ehsan Zabardast, Julian Frattini, Javier Gonzalez-Huerta, Daniel Mendez, Tony Gorschek, Krzysztof Wnuk
Assets in Software Engineering: What are they after all?
Manuscript submitted to the Journal of Systems and Software
null
10.1016/j.jss.2022.111485
null
cs.SE
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
During the development and maintenance of software-intensive products or services, we depend on various artefacts. Some of those artefacts, we deem central to the feasibility of a project and the product's final quality. Typically, these central artefacts are referred to as assets. However, despite their central role in the software development process, little thought is yet invested into what eventually characterises as an asset, often resulting in many terms and underlying concepts being mixed and used inconsistently. A precise terminology of assets and related concepts, such as asset degradation, are crucial for setting up a new generation of cost-effective software engineering practices. In this position paper, we critically reflect upon the notion of assets in software engineering. As a starting point, we define the terminology and concepts of assets and extend the reasoning behind them. We explore assets' characteristics and discuss what asset degradation is as well as its various types and the implications that asset degradation might bring for the planning, realisation, and evolution of software-intensive products and services over time. We aspire to contribute to a more standardised definition of assets in software engineering and foster research endeavours and their practical dissemination in a common, more unified direction.
[ { "created": "Tue, 19 Jan 2021 18:31:33 GMT", "version": "v1" }, { "created": "Sun, 24 Oct 2021 13:19:22 GMT", "version": "v2" }, { "created": "Wed, 23 Feb 2022 14:32:15 GMT", "version": "v3" }, { "created": "Wed, 11 May 2022 21:31:28 GMT", "version": "v4" }, { "created": "Mon, 11 Jul 2022 14:57:38 GMT", "version": "v5" } ]
2022-08-29
[ [ "Zabardast", "Ehsan", "" ], [ "Frattini", "Julian", "" ], [ "Gonzalez-Huerta", "Javier", "" ], [ "Mendez", "Daniel", "" ], [ "Gorschek", "Tony", "" ], [ "Wnuk", "Krzysztof", "" ] ]
During the development and maintenance of software-intensive products or services, we depend on various artefacts. Some of those artefacts, we deem central to the feasibility of a project and the product's final quality. Typically, these central artefacts are referred to as assets. However, despite their central role in the software development process, little thought is yet invested into what eventually characterises as an asset, often resulting in many terms and underlying concepts being mixed and used inconsistently. A precise terminology of assets and related concepts, such as asset degradation, are crucial for setting up a new generation of cost-effective software engineering practices. In this position paper, we critically reflect upon the notion of assets in software engineering. As a starting point, we define the terminology and concepts of assets and extend the reasoning behind them. We explore assets' characteristics and discuss what asset degradation is as well as its various types and the implications that asset degradation might bring for the planning, realisation, and evolution of software-intensive products and services over time. We aspire to contribute to a more standardised definition of assets in software engineering and foster research endeavours and their practical dissemination in a common, more unified direction.
2108.02290
Max Willsey
Yihong Zhang, Yisu Remy Wang, Max Willsey, Zachary Tatlock
Relational E-Matching
POPL 2022
null
null
null
cs.DB cs.PL
http://creativecommons.org/licenses/by/4.0/
We present a new approach to e-matching based on relational join; in particular, we apply recent database query execution techniques to guarantee worst-case optimal run time. Compared to the conventional backtracking approach that always searches the e-graph "top down", our new relational e-matching approach can better exploit pattern structure by searching the e-graph according to an optimized query plan. We also establish the first data complexity result for e-matching, bounding run time as a function of the e-graph size and output size. We prototyped and evaluated our technique in the state-of-the-art egg e-graph framework. Compared to a conventional baseline, relational e-matching is simpler to implement and orders of magnitude faster in practice.
[ { "created": "Wed, 4 Aug 2021 21:23:28 GMT", "version": "v1" }, { "created": "Wed, 5 Jan 2022 21:21:44 GMT", "version": "v2" } ]
2022-01-07
[ [ "Zhang", "Yihong", "" ], [ "Wang", "Yisu Remy", "" ], [ "Willsey", "Max", "" ], [ "Tatlock", "Zachary", "" ] ]
We present a new approach to e-matching based on relational join; in particular, we apply recent database query execution techniques to guarantee worst-case optimal run time. Compared to the conventional backtracking approach that always searches the e-graph "top down", our new relational e-matching approach can better exploit pattern structure by searching the e-graph according to an optimized query plan. We also establish the first data complexity result for e-matching, bounding run time as a function of the e-graph size and output size. We prototyped and evaluated our technique in the state-of-the-art egg e-graph framework. Compared to a conventional baseline, relational e-matching is simpler to implement and orders of magnitude faster in practice.
1907.09247
Jorge Fandinno
Jorge Fandinno
Founded (Auto)Epistemic Equilibrium Logic Satisfies Epistemic Splitting
Paper presented at the 35th International Conference on Logic Programming (ICLP 2019), Las Cruces, New Mexico, USA, 20-25 September 2019, 16 pages
Theory and Practice of Logic Programming 19 (2019) 671-687
10.1017/S1471068419000127
null
cs.LO cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In a recent line of research, two familiar concepts from logic programming semantics (unfounded sets and splitting) were extrapolated to the case of epistemic logic programs. The property of epistemic splitting provides a natural and modular way to understand programs without epistemic cycles but, surprisingly, was only fulfilled by Gelfond's original semantics (G91), among the many proposals in the literature. On the other hand, G91 may suffer from a kind of self-supported, unfounded derivations when epistemic cycles come into play. Recently, the absence of these derivations was also formalised as a property of epistemic semantics called foundedness. Moreover, a first semantics proved to satisfy foundedness was also proposed, the so-called Founded Autoepistemic Equilibrium Logic (FAEEL). In this paper, we prove that FAEEL also satisfies the epistemic splitting property something that, together with foundedness, was not fulfilled by any other approach up to date. To prove this result, we provide an alternative characterisation of FAEEL as a combination of G91 with a simpler logic we called Founded Epistemic Equilibrium Logic (FEEL), which is somehow an extrapolation of the stable model semantics to the modal logic S5. Under consideration for acceptance in TPLP.
[ { "created": "Mon, 22 Jul 2019 11:48:15 GMT", "version": "v1" }, { "created": "Thu, 20 Feb 2020 15:33:26 GMT", "version": "v2" } ]
2020-02-21
[ [ "Fandinno", "Jorge", "" ] ]
In a recent line of research, two familiar concepts from logic programming semantics (unfounded sets and splitting) were extrapolated to the case of epistemic logic programs. The property of epistemic splitting provides a natural and modular way to understand programs without epistemic cycles but, surprisingly, was only fulfilled by Gelfond's original semantics (G91), among the many proposals in the literature. On the other hand, G91 may suffer from a kind of self-supported, unfounded derivations when epistemic cycles come into play. Recently, the absence of these derivations was also formalised as a property of epistemic semantics called foundedness. Moreover, a first semantics proved to satisfy foundedness was also proposed, the so-called Founded Autoepistemic Equilibrium Logic (FAEEL). In this paper, we prove that FAEEL also satisfies the epistemic splitting property something that, together with foundedness, was not fulfilled by any other approach up to date. To prove this result, we provide an alternative characterisation of FAEEL as a combination of G91 with a simpler logic we called Founded Epistemic Equilibrium Logic (FEEL), which is somehow an extrapolation of the stable model semantics to the modal logic S5. Under consideration for acceptance in TPLP.
1902.05260
Peng Wang
Peng Wang, Hong Xu, Xin Jin, Tao Wang
Flash: Efficient Dynamic Routing for Offchain Networks
null
null
null
null
cs.NI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Offchain networks emerge as a promising solution to address the scalability challenge of blockchain. Participants directly make payments through a network of payment channels without the overhead of committing onchain transactions. Routing is critical to the performance of offchain networks. Existing solutions use either static routing with poor performance or dynamic routing with high overhead to obtain the dynamic channel balance information. In this paper, we propose Flash, a new dynamic routing solution that leverages the unique characteristics of transactions in offchain networks to strike a better tradeoff between path optimality and probing overhead. By studying the traces of real offchain networks, we find that the payment sizes are heavy-tailed, and most payments are highly recurrent. Flash thus differentiates the treatment of elephant payments from mice payments. It uses a modified max-flow algorithm for elephant payments to find paths with sufficient capacity, and strategically routes the payment across paths to minimize the transaction fees. Mice payments are directly sent by looking up a routing table with a few precomputed paths to reduce probing overhead. Testbed experiments and data-driven simulations show that Flash improves the success volume of payments by up to 2.3x compared to the state-of-the-art routing algorithm.
[ { "created": "Thu, 14 Feb 2019 08:36:57 GMT", "version": "v1" }, { "created": "Tue, 11 Jun 2019 03:01:10 GMT", "version": "v2" } ]
2019-06-12
[ [ "Wang", "Peng", "" ], [ "Xu", "Hong", "" ], [ "Jin", "Xin", "" ], [ "Wang", "Tao", "" ] ]
Offchain networks emerge as a promising solution to address the scalability challenge of blockchain. Participants directly make payments through a network of payment channels without the overhead of committing onchain transactions. Routing is critical to the performance of offchain networks. Existing solutions use either static routing with poor performance or dynamic routing with high overhead to obtain the dynamic channel balance information. In this paper, we propose Flash, a new dynamic routing solution that leverages the unique characteristics of transactions in offchain networks to strike a better tradeoff between path optimality and probing overhead. By studying the traces of real offchain networks, we find that the payment sizes are heavy-tailed, and most payments are highly recurrent. Flash thus differentiates the treatment of elephant payments from mice payments. It uses a modified max-flow algorithm for elephant payments to find paths with sufficient capacity, and strategically routes the payment across paths to minimize the transaction fees. Mice payments are directly sent by looking up a routing table with a few precomputed paths to reduce probing overhead. Testbed experiments and data-driven simulations show that Flash improves the success volume of payments by up to 2.3x compared to the state-of-the-art routing algorithm.
1811.05932
Xi Liu
Xi Liu, Ping-Chun Hsieh, Nick Duffield, Rui Chen, Muhe Xie, Xidao Wen
Streaming Network Embedding through Local Actions
null
null
null
null
cs.LG cs.SI stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Recently, considerable research attention has been paid to network embedding, a popular approach to construct feature vectors of vertices. Due to the curse of dimensionality and sparsity in graphical datasets, this approach has become indispensable for machine learning tasks over large networks. The majority of existing literature has considered this technique under the assumption that the network is static. However, networks in many applications, nodes and edges accrue to a growing network as a streaming. A small number of very recent results have addressed the problem of embedding for dynamic networks. However, they either rely on knowledge of vertex attributes, suffer high-time complexity or need to be re-trained without closed-form expression. Thus the approach of adapting the existing methods to the streaming environment faces non-trivial technical challenges. These challenges motivate developing new approaches to the problems of streaming network embedding. In this paper, We propose a new framework that is able to generate latent features for new vertices with high efficiency and low complexity under specified iteration rounds. We formulate a constrained optimization problem for the modification of the representation resulting from a stream arrival. We show this problem has no closed-form solution and instead develop an online approximation solution. Our solution follows three steps: (1) identify vertices affected by new vertices, (2) generate latent features for new vertices, and (3) update the latent features of the most affected vertices. The generated representations are provably feasible and not far from the optimal ones in terms of expectation. Multi-class classification and clustering on five real-world networks demonstrate that our model can efficiently update vertex representations and simultaneously achieve comparable or even better performance.
[ { "created": "Wed, 14 Nov 2018 18:02:29 GMT", "version": "v1" } ]
2018-11-15
[ [ "Liu", "Xi", "" ], [ "Hsieh", "Ping-Chun", "" ], [ "Duffield", "Nick", "" ], [ "Chen", "Rui", "" ], [ "Xie", "Muhe", "" ], [ "Wen", "Xidao", "" ] ]
Recently, considerable research attention has been paid to network embedding, a popular approach to construct feature vectors of vertices. Due to the curse of dimensionality and sparsity in graphical datasets, this approach has become indispensable for machine learning tasks over large networks. The majority of existing literature has considered this technique under the assumption that the network is static. However, networks in many applications, nodes and edges accrue to a growing network as a streaming. A small number of very recent results have addressed the problem of embedding for dynamic networks. However, they either rely on knowledge of vertex attributes, suffer high-time complexity or need to be re-trained without closed-form expression. Thus the approach of adapting the existing methods to the streaming environment faces non-trivial technical challenges. These challenges motivate developing new approaches to the problems of streaming network embedding. In this paper, We propose a new framework that is able to generate latent features for new vertices with high efficiency and low complexity under specified iteration rounds. We formulate a constrained optimization problem for the modification of the representation resulting from a stream arrival. We show this problem has no closed-form solution and instead develop an online approximation solution. Our solution follows three steps: (1) identify vertices affected by new vertices, (2) generate latent features for new vertices, and (3) update the latent features of the most affected vertices. The generated representations are provably feasible and not far from the optimal ones in terms of expectation. Multi-class classification and clustering on five real-world networks demonstrate that our model can efficiently update vertex representations and simultaneously achieve comparable or even better performance.
2304.07954
Jihao Huang
Jihao Huang, Jun Zeng, Xuemin Chi, Koushil Sreenath, Zhitao Liu and Hongye Su
Velocity Obstacle for Polytopic Collision Avoidance for Distributed Multi-robot Systems
Accepted to IEEE Robotics and Automation Letters (RA-L) 2023, with open source repository released
null
null
null
cs.RO cs.SY eess.SY math.OC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Obstacle avoidance for multi-robot navigation with polytopic shapes is challenging. Existing works simplify the system dynamics or consider it as a convex or non-convex optimization problem with positive distance constraints between robots, which limits real-time performance and scalability. Additionally, generating collision-free behavior for polytopic-shaped robots is harder due to implicit and non-differentiable distance functions between polytopes. In this paper, we extend the concept of velocity obstacle (VO) principle for polytopic-shaped robots and propose a novel approach to construct the VO in the function of vertex coordinates and other robot's states. Compared with existing work about obstacle avoidance between polytopic-shaped robots, our approach is much more computationally efficient as the proposed approach for construction of VO between polytopes is optimization-free. Based on VO representation for polytopic shapes, we later propose a navigation approach for distributed multi-robot systems. We validate our proposed VO representation and navigation approach in multiple challenging scenarios including large-scale randomized tests, and our approach outperforms the state of art in many evaluation metrics, including completion rate, deadlock rate, and the average travel distance.
[ { "created": "Mon, 17 Apr 2023 02:42:48 GMT", "version": "v1" }, { "created": "Mon, 10 Jun 2024 04:40:33 GMT", "version": "v2" } ]
2024-06-11
[ [ "Huang", "Jihao", "" ], [ "Zeng", "Jun", "" ], [ "Chi", "Xuemin", "" ], [ "Sreenath", "Koushil", "" ], [ "Liu", "Zhitao", "" ], [ "Su", "Hongye", "" ] ]
Obstacle avoidance for multi-robot navigation with polytopic shapes is challenging. Existing works simplify the system dynamics or consider it as a convex or non-convex optimization problem with positive distance constraints between robots, which limits real-time performance and scalability. Additionally, generating collision-free behavior for polytopic-shaped robots is harder due to implicit and non-differentiable distance functions between polytopes. In this paper, we extend the concept of velocity obstacle (VO) principle for polytopic-shaped robots and propose a novel approach to construct the VO in the function of vertex coordinates and other robot's states. Compared with existing work about obstacle avoidance between polytopic-shaped robots, our approach is much more computationally efficient as the proposed approach for construction of VO between polytopes is optimization-free. Based on VO representation for polytopic shapes, we later propose a navigation approach for distributed multi-robot systems. We validate our proposed VO representation and navigation approach in multiple challenging scenarios including large-scale randomized tests, and our approach outperforms the state of art in many evaluation metrics, including completion rate, deadlock rate, and the average travel distance.
1907.13432
Dong Liu
Dong Liu, Minh Th\`anh Vu, Saikat Chatterjee, Lars K. Rasmussen
Neural Network based Explicit Mixture Models and Expectation-maximization based Learning
IJCNN 2020
null
null
null
cs.LG stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
We propose two neural network based mixture models in this article. The proposed mixture models are explicit in nature. The explicit models have analytical forms with the advantages of computing likelihood and efficiency of generating samples. Computation of likelihood is an important aspect of our models. Expectation-maximization based algorithms are developed for learning parameters of the proposed models. We provide sufficient conditions to realize the expectation-maximization based learning. The main requirements are invertibility of neural networks that are used as generators and Jacobian computation of functional form of the neural networks. The requirements are practically realized using a flow-based neural network. In our first mixture model, we use multiple flow-based neural networks as generators. Naturally the model is complex. A single latent variable is used as the common input to all the neural networks. The second mixture model uses a single flow-based neural network as a generator to reduce complexity. The single generator has a latent variable input that follows a Gaussian mixture distribution. We demonstrate efficiency of proposed mixture models through extensive experiments for generating samples and maximum likelihood based classification.
[ { "created": "Wed, 31 Jul 2019 11:57:17 GMT", "version": "v1" }, { "created": "Sun, 24 May 2020 19:57:55 GMT", "version": "v2" } ]
2020-05-26
[ [ "Liu", "Dong", "" ], [ "Vu", "Minh Thành", "" ], [ "Chatterjee", "Saikat", "" ], [ "Rasmussen", "Lars K.", "" ] ]
We propose two neural network based mixture models in this article. The proposed mixture models are explicit in nature. The explicit models have analytical forms with the advantages of computing likelihood and efficiency of generating samples. Computation of likelihood is an important aspect of our models. Expectation-maximization based algorithms are developed for learning parameters of the proposed models. We provide sufficient conditions to realize the expectation-maximization based learning. The main requirements are invertibility of neural networks that are used as generators and Jacobian computation of functional form of the neural networks. The requirements are practically realized using a flow-based neural network. In our first mixture model, we use multiple flow-based neural networks as generators. Naturally the model is complex. A single latent variable is used as the common input to all the neural networks. The second mixture model uses a single flow-based neural network as a generator to reduce complexity. The single generator has a latent variable input that follows a Gaussian mixture distribution. We demonstrate efficiency of proposed mixture models through extensive experiments for generating samples and maximum likelihood based classification.
2009.10233
Boyuan Feng
Boyuan Feng, Yuke Wang, Xu Li, and Yufei Ding
Scalable Adversarial Attack on Graph Neural Networks with Alternating Direction Method of Multipliers
null
null
null
null
cs.LG cs.AI stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Graph neural networks (GNNs) have achieved high performance in analyzing graph-structured data and have been widely deployed in safety-critical areas, such as finance and autonomous driving. However, only a few works have explored GNNs' robustness to adversarial attacks, and their designs are usually limited by the scale of input datasets (i.e., focusing on small graphs with only thousands of nodes). In this work, we propose, SAG, the first scalable adversarial attack method with Alternating Direction Method of Multipliers (ADMM). We first decouple the large-scale graph into several smaller graph partitions and cast the original problem into several subproblems. Then, we propose to solve these subproblems using projected gradient descent on both the graph topology and the node features that lead to considerably lower memory consumption compared to the conventional attack methods. Rigorous experiments further demonstrate that SAG can significantly reduce the computation and memory overhead compared with the state-of-the-art approach, making SAG applicable towards graphs with large size of nodes and edges.
[ { "created": "Tue, 22 Sep 2020 00:33:36 GMT", "version": "v1" } ]
2020-09-23
[ [ "Feng", "Boyuan", "" ], [ "Wang", "Yuke", "" ], [ "Li", "Xu", "" ], [ "Ding", "Yufei", "" ] ]
Graph neural networks (GNNs) have achieved high performance in analyzing graph-structured data and have been widely deployed in safety-critical areas, such as finance and autonomous driving. However, only a few works have explored GNNs' robustness to adversarial attacks, and their designs are usually limited by the scale of input datasets (i.e., focusing on small graphs with only thousands of nodes). In this work, we propose, SAG, the first scalable adversarial attack method with Alternating Direction Method of Multipliers (ADMM). We first decouple the large-scale graph into several smaller graph partitions and cast the original problem into several subproblems. Then, we propose to solve these subproblems using projected gradient descent on both the graph topology and the node features that lead to considerably lower memory consumption compared to the conventional attack methods. Rigorous experiments further demonstrate that SAG can significantly reduce the computation and memory overhead compared with the state-of-the-art approach, making SAG applicable towards graphs with large size of nodes and edges.
2407.17483
Sherri WeitlHarms
Sherri Harms
Tackling CS education in K-12: Implementing a Google CS4HS Grant Program in a Rural Underserved Area
Presented at and published in the proceedings of MICS 2017: https://www.micsymposium.org/mics_2017_proceedings/docs/MICS_2017_paper_44.pdf; 8 pages, 2 figures
Harms, S. K. (2017). Tackling CS education in K-12: Implementing a Google CS4HS Grant Program. La Crosse, WI: 2017 Midwest Instructional Computing Symposium Proceedings
null
null
cs.CY
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Providing computer science (CS) offerings in the K-12 education system is often limited by the lack of experienced teachers, especially in small or rural underserved school districts. By helping teachers in underserved areas develop CS curriculum and helping them become certified to teach CS courses, more young people in underserved areas are aware of IT-career opportunities, and prepared for CS education at the university level, which ultimately helps tackle the IT workforce deficit in the United States. This paper discusses a successful implementation of a Google CS4HS grant to a rural underserved area, as well as lessons learned through the implementation of the program. Key elements in the implementation included a face-to-face hands-on workshop, followed by a seven week graduate-level online summer course for the teachers to learn and develop curriculum that covers the CS concepts they will be teaching. The teachers were supported with an online community of practice for the year as they implemented the curriculum.
[ { "created": "Tue, 2 Jul 2024 18:36:06 GMT", "version": "v1" } ]
2024-07-26
[ [ "Harms", "Sherri", "" ] ]
Providing computer science (CS) offerings in the K-12 education system is often limited by the lack of experienced teachers, especially in small or rural underserved school districts. By helping teachers in underserved areas develop CS curriculum and helping them become certified to teach CS courses, more young people in underserved areas are aware of IT-career opportunities, and prepared for CS education at the university level, which ultimately helps tackle the IT workforce deficit in the United States. This paper discusses a successful implementation of a Google CS4HS grant to a rural underserved area, as well as lessons learned through the implementation of the program. Key elements in the implementation included a face-to-face hands-on workshop, followed by a seven week graduate-level online summer course for the teachers to learn and develop curriculum that covers the CS concepts they will be teaching. The teachers were supported with an online community of practice for the year as they implemented the curriculum.
2101.11946
Stefan Heidekr\"uger
Stefan Heidekr\"uger, Paul Sutterer, Nils Kohring, Maximilian Fichtl, and Martin Bichler
Equilibrium Learning in Combinatorial Auctions: Computing Approximate Bayesian Nash Equilibria via Pseudogradient Dynamics
This version includes the supplementary material with additional proofs
null
null
null
cs.GT cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Applications of combinatorial auctions (CA) as market mechanisms are prevalent in practice, yet their Bayesian Nash equilibria (BNE) remain poorly understood. Analytical solutions are known only for a few cases where the problem can be reformulated as a tractable partial differential equation (PDE). In the general case, finding BNE is known to be computationally hard. Previous work on numerical computation of BNE in auctions has relied either on solving such PDEs explicitly, calculating pointwise best-responses in strategy space, or iteratively solving restricted subgames. In this study, we present a generic yet scalable alternative multi-agent equilibrium learning method that represents strategies as neural networks and applies policy iteration based on gradient dynamics in self-play. Most auctions are ex-post nondifferentiable, so gradients may be unavailable or misleading, and we rely on suitable pseudogradient estimates instead. Although it is well-known that gradient dynamics cannot guarantee convergence to NE in general, we observe fast and robust convergence to approximate BNE in a wide variety of auctions and present a sufficient condition for convergence
[ { "created": "Thu, 28 Jan 2021 11:53:32 GMT", "version": "v1" }, { "created": "Sat, 6 Feb 2021 11:47:52 GMT", "version": "v2" } ]
2021-02-09
[ [ "Heidekrüger", "Stefan", "" ], [ "Sutterer", "Paul", "" ], [ "Kohring", "Nils", "" ], [ "Fichtl", "Maximilian", "" ], [ "Bichler", "Martin", "" ] ]
Applications of combinatorial auctions (CA) as market mechanisms are prevalent in practice, yet their Bayesian Nash equilibria (BNE) remain poorly understood. Analytical solutions are known only for a few cases where the problem can be reformulated as a tractable partial differential equation (PDE). In the general case, finding BNE is known to be computationally hard. Previous work on numerical computation of BNE in auctions has relied either on solving such PDEs explicitly, calculating pointwise best-responses in strategy space, or iteratively solving restricted subgames. In this study, we present a generic yet scalable alternative multi-agent equilibrium learning method that represents strategies as neural networks and applies policy iteration based on gradient dynamics in self-play. Most auctions are ex-post nondifferentiable, so gradients may be unavailable or misleading, and we rely on suitable pseudogradient estimates instead. Although it is well-known that gradient dynamics cannot guarantee convergence to NE in general, we observe fast and robust convergence to approximate BNE in a wide variety of auctions and present a sufficient condition for convergence
2205.11020
Rohitash Chandra
Rohitash Chandra, Mukul Ranjan
Artificial intelligence for topic modelling in Hindu philosophy: mapping themes between the Upanishads and the Bhagavad Gita
null
null
10.1371/journal.pone.0273476
null
cs.CL
http://creativecommons.org/licenses/by/4.0/
A distinct feature of Hindu religious and philosophical text is that they come from a library of texts rather than single source. The Upanishads is known as one of the oldest philosophical texts in the world that forms the foundation of Hindu philosophy. The Bhagavad Gita is core text of Hindu philosophy and is known as a text that summarises the key philosophies of the Upanishads with major focus on the philosophy of karma. These texts have been translated into many languages and there exists studies about themes and topics that are prominent; however, there is not much study of topic modelling using language models which are powered by deep learning. In this paper, we use advanced language produces such as BERT to provide topic modelling of the key texts of the Upanishads and the Bhagavad Gita. We analyse the distinct and overlapping topics amongst the texts and visualise the link of selected texts of the Upanishads with Bhagavad Gita. Our results show a very high similarity between the topics of these two texts with the mean cosine similarity of 73%. We find that out of the fourteen topics extracted from the Bhagavad Gita, nine of them have a cosine similarity of more than 70% with the topics of the Upanishads. We also found that topics generated by the BERT-based models show very high coherence as compared to that of conventional models. Our best performing model gives a coherence score of 73% on the Bhagavad Gita and 69% on The Upanishads. The visualization of the low dimensional embeddings of these texts shows very clear overlapping among their topics adding another level of validation to our results.
[ { "created": "Mon, 23 May 2022 03:39:00 GMT", "version": "v1" } ]
2022-10-12
[ [ "Chandra", "Rohitash", "" ], [ "Ranjan", "Mukul", "" ] ]
A distinct feature of Hindu religious and philosophical text is that they come from a library of texts rather than single source. The Upanishads is known as one of the oldest philosophical texts in the world that forms the foundation of Hindu philosophy. The Bhagavad Gita is core text of Hindu philosophy and is known as a text that summarises the key philosophies of the Upanishads with major focus on the philosophy of karma. These texts have been translated into many languages and there exists studies about themes and topics that are prominent; however, there is not much study of topic modelling using language models which are powered by deep learning. In this paper, we use advanced language produces such as BERT to provide topic modelling of the key texts of the Upanishads and the Bhagavad Gita. We analyse the distinct and overlapping topics amongst the texts and visualise the link of selected texts of the Upanishads with Bhagavad Gita. Our results show a very high similarity between the topics of these two texts with the mean cosine similarity of 73%. We find that out of the fourteen topics extracted from the Bhagavad Gita, nine of them have a cosine similarity of more than 70% with the topics of the Upanishads. We also found that topics generated by the BERT-based models show very high coherence as compared to that of conventional models. Our best performing model gives a coherence score of 73% on the Bhagavad Gita and 69% on The Upanishads. The visualization of the low dimensional embeddings of these texts shows very clear overlapping among their topics adding another level of validation to our results.
1711.02447
Kaushik Sarker
Md. Maruf Hassan, Kaushik Sarker, Saikat Biswas, Md. Hasan Sharif
Detection of Wordpress Content Injection Vulnerability
null
null
10.5121/ijci.2017.6501
null
cs.CR
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The popularity of content management software (CMS) is growing vastly to the web developers and the business people because of its capacity for easy accessibility, manageability and usability of the distributed website contents. As per the statistics of Built with, 32% of the web applications are developed with WordPress(WP) among all other CMSs [1]. It is obvious that quite a good number of web applications were built with WP in version 4.7.0 and 4.7.1. A recent research reveals that content injection vulnerability was found available in the above two versions of WP [2]. Unauthorized content injection by an intruder in a CMS managed application is one of the serious problems for the business as well as for the web owner.Therefore, detection of the vulnerability becomes a critical issue for this time. In this paper, we have discussed about the root cause of WP content injection of the above versions and have also proposed a detection model for the given vulnerability. A tool, SAISAN has been implemented as per our anticipated model and conducted an examination on 176 WP developed web applications using SAISAN. We achieved the accuracy of 92% of the result of SAISAN as compared to manual black box testing outcome.
[ { "created": "Tue, 7 Nov 2017 13:01:18 GMT", "version": "v1" } ]
2017-11-08
[ [ "Hassan", "Md. Maruf", "" ], [ "Sarker", "Kaushik", "" ], [ "Biswas", "Saikat", "" ], [ "Sharif", "Md. Hasan", "" ] ]
The popularity of content management software (CMS) is growing vastly to the web developers and the business people because of its capacity for easy accessibility, manageability and usability of the distributed website contents. As per the statistics of Built with, 32% of the web applications are developed with WordPress(WP) among all other CMSs [1]. It is obvious that quite a good number of web applications were built with WP in version 4.7.0 and 4.7.1. A recent research reveals that content injection vulnerability was found available in the above two versions of WP [2]. Unauthorized content injection by an intruder in a CMS managed application is one of the serious problems for the business as well as for the web owner.Therefore, detection of the vulnerability becomes a critical issue for this time. In this paper, we have discussed about the root cause of WP content injection of the above versions and have also proposed a detection model for the given vulnerability. A tool, SAISAN has been implemented as per our anticipated model and conducted an examination on 176 WP developed web applications using SAISAN. We achieved the accuracy of 92% of the result of SAISAN as compared to manual black box testing outcome.
1709.00700
Sebastian Bre{\ss}
Sebastian Bre{\ss} and Bastian K\"ocher and Henning Funke and Tilmann Rabl and Volker Markl
Generating Custom Code for Efficient Query Execution on Heterogeneous Processors
22 pages
null
null
null
cs.DB cs.DC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Processor manufacturers build increasingly specialized processors to mitigate the effects of the power wall to deliver improved performance. Currently, database engines are manually optimized for each processor: A costly and error prone process. In this paper, we propose concepts to enable the database engine to perform per-processor optimization automatically. Our core idea is to create variants of generated code and to learn a fast variant for each processor. We create variants by modifying parallelization strategies, specializing data structures, and applying different code transformations. Our experimental results show that the performance of variants may diverge up to two orders of magnitude. Therefore, we need to generate custom code for each processor to achieve peak performance. We show that our approach finds a fast custom variant for multi-core CPUs, GPUs, and MICs.
[ { "created": "Sun, 3 Sep 2017 11:16:31 GMT", "version": "v1" } ]
2017-09-05
[ [ "Breß", "Sebastian", "" ], [ "Köcher", "Bastian", "" ], [ "Funke", "Henning", "" ], [ "Rabl", "Tilmann", "" ], [ "Markl", "Volker", "" ] ]
Processor manufacturers build increasingly specialized processors to mitigate the effects of the power wall to deliver improved performance. Currently, database engines are manually optimized for each processor: A costly and error prone process. In this paper, we propose concepts to enable the database engine to perform per-processor optimization automatically. Our core idea is to create variants of generated code and to learn a fast variant for each processor. We create variants by modifying parallelization strategies, specializing data structures, and applying different code transformations. Our experimental results show that the performance of variants may diverge up to two orders of magnitude. Therefore, we need to generate custom code for each processor to achieve peak performance. We show that our approach finds a fast custom variant for multi-core CPUs, GPUs, and MICs.
1604.08552
Sameh Sorour
Rabe Arshad, Hesham ElSawy, Sameh Sorour, Tareq Y. Al-Naffouri, Mohamed-Slim Alouini
Handover Management in Dense Cellular Networks: A Stochastic Geometry Approach
7 pages, 7 figures, ICC 2016
null
null
null
cs.NI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Cellular operators are continuously densifying their networks to cope with the ever-increasing capacity demand. Furthermore, an extreme densification phase for cellular networks is foreseen to fulfill the ambitious fifth generation (5G) performance requirements. Network densification improves spectrum utilization and network capacity by shrinking base stations' (BSs) footprints and reusing the same spectrum more frequently over the spatial domain. However, network densification also increases the handover (HO) rate, which may diminish the capacity gains for mobile users due to HO delays. In highly dense 5G cellular networks, HO delays may neutralize or even negate the gains offered by network densification. In this paper, we present an analytical paradigm, based on stochastic geometry, to quantify the effect of HO delay on the average user rate in cellular networks. To this end, we propose a flexible handover scheme to reduce HO delay in case of highly dense cellular networks. This scheme allows skipping the HO procedure with some BSs along users' trajectories. The performance evaluation and testing of this scheme for only single HO skipping shows considerable gains in many practical scenarios.
[ { "created": "Thu, 28 Apr 2016 18:44:12 GMT", "version": "v1" } ]
2016-04-29
[ [ "Arshad", "Rabe", "" ], [ "ElSawy", "Hesham", "" ], [ "Sorour", "Sameh", "" ], [ "Al-Naffouri", "Tareq Y.", "" ], [ "Alouini", "Mohamed-Slim", "" ] ]
Cellular operators are continuously densifying their networks to cope with the ever-increasing capacity demand. Furthermore, an extreme densification phase for cellular networks is foreseen to fulfill the ambitious fifth generation (5G) performance requirements. Network densification improves spectrum utilization and network capacity by shrinking base stations' (BSs) footprints and reusing the same spectrum more frequently over the spatial domain. However, network densification also increases the handover (HO) rate, which may diminish the capacity gains for mobile users due to HO delays. In highly dense 5G cellular networks, HO delays may neutralize or even negate the gains offered by network densification. In this paper, we present an analytical paradigm, based on stochastic geometry, to quantify the effect of HO delay on the average user rate in cellular networks. To this end, we propose a flexible handover scheme to reduce HO delay in case of highly dense cellular networks. This scheme allows skipping the HO procedure with some BSs along users' trajectories. The performance evaluation and testing of this scheme for only single HO skipping shows considerable gains in many practical scenarios.
1903.08314
Shigeru Furuichi Dr.
Shigeru Furuichi and Nicu\c{s}or Minculete
Inequalities related to some types of entropies and divergences
21 pages
null
10.1016/j.physa.2019.121907
null
cs.IT math.CA math.IT
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
The aim of this paper is to discuss new results concerning some kinds of parametric extended entropies and divergences. As a result of our studies for mathematical properties on entropy and divergence, we give new bounds for the Tsallis quasilinear entropy and divergence by applying the Hermite-Hadamard inequality. We also give bounds for biparametrical extended entropies and divergences which have been given in \cite{7}. In addition, we study $(r,q)$-quasilinear entropies and divergences as alternative biparametrical extended entropy and divergence, and then we give bounds for them. Finally we obtain inequalities for an extended Lin's divergence and some characterizations of Fermi-Dirac entropy and Bose-Einstein entropy.
[ { "created": "Wed, 20 Mar 2019 02:03:40 GMT", "version": "v1" }, { "created": "Sat, 13 Jul 2019 02:38:55 GMT", "version": "v2" } ]
2019-07-24
[ [ "Furuichi", "Shigeru", "" ], [ "Minculete", "Nicuşor", "" ] ]
The aim of this paper is to discuss new results concerning some kinds of parametric extended entropies and divergences. As a result of our studies for mathematical properties on entropy and divergence, we give new bounds for the Tsallis quasilinear entropy and divergence by applying the Hermite-Hadamard inequality. We also give bounds for biparametrical extended entropies and divergences which have been given in \cite{7}. In addition, we study $(r,q)$-quasilinear entropies and divergences as alternative biparametrical extended entropy and divergence, and then we give bounds for them. Finally we obtain inequalities for an extended Lin's divergence and some characterizations of Fermi-Dirac entropy and Bose-Einstein entropy.
2012.01227
Taehyeong Kim
Taehyeong Kim, Injune Hwang, Hyundo Lee, Hyunseo Kim, Won-Seok Choi, Joseph J. Lim, Byoung-Tak Zhang
Message Passing Adaptive Resonance Theory for Online Active Semi-supervised Learning
Accepted to ICML 2021
null
null
null
cs.LG cs.AI
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Active learning is widely used to reduce labeling effort and training time by repeatedly querying only the most beneficial samples from unlabeled data. In real-world problems where data cannot be stored indefinitely due to limited storage or privacy issues, the query selection and the model update should be performed as soon as a new data sample is observed. Various online active learning methods have been studied to deal with these challenges; however, there are difficulties in selecting representative query samples and updating the model efficiently without forgetting. In this study, we propose Message Passing Adaptive Resonance Theory (MPART) that learns the distribution and topology of input data online. Through message passing on the topological graph, MPART actively queries informative and representative samples, and continuously improves the classification performance using both labeled and unlabeled data. We evaluate our model in stream-based selective sampling scenarios with comparable query selection strategies, showing that MPART significantly outperforms competitive models.
[ { "created": "Wed, 2 Dec 2020 14:14:42 GMT", "version": "v1" }, { "created": "Wed, 24 Feb 2021 10:04:51 GMT", "version": "v2" }, { "created": "Sat, 10 Jul 2021 05:58:30 GMT", "version": "v3" } ]
2021-07-13
[ [ "Kim", "Taehyeong", "" ], [ "Hwang", "Injune", "" ], [ "Lee", "Hyundo", "" ], [ "Kim", "Hyunseo", "" ], [ "Choi", "Won-Seok", "" ], [ "Lim", "Joseph J.", "" ], [ "Zhang", "Byoung-Tak", "" ] ]
Active learning is widely used to reduce labeling effort and training time by repeatedly querying only the most beneficial samples from unlabeled data. In real-world problems where data cannot be stored indefinitely due to limited storage or privacy issues, the query selection and the model update should be performed as soon as a new data sample is observed. Various online active learning methods have been studied to deal with these challenges; however, there are difficulties in selecting representative query samples and updating the model efficiently without forgetting. In this study, we propose Message Passing Adaptive Resonance Theory (MPART) that learns the distribution and topology of input data online. Through message passing on the topological graph, MPART actively queries informative and representative samples, and continuously improves the classification performance using both labeled and unlabeled data. We evaluate our model in stream-based selective sampling scenarios with comparable query selection strategies, showing that MPART significantly outperforms competitive models.
cs/0103008
Ke Xu
Shilong Ma, Yuefei Sui, Ke Xu
The Limits of Horn Logic Programs
11 pages, added new results. Welcome any comments to kexu@nlsde.buaa.edu.cn
In P. J. Stuckey (Ed.): Proc. of 18th ICLP (short paper), LNCS 2401, p. 467, Denmark, 2002.
null
null
cs.LO cs.PL
null
Given a sequence $\{\Pi_n\}$ of Horn logic programs, the limit $\Pi$ of $\{\Pi_n\}$ is the set of the clauses such that every clause in $\Pi$ belongs to almost every $\Pi_n$ and every clause in infinitely many $\Pi_n$'s belongs to $\Pi$ also. The limit program $\Pi$ is still Horn but may be infinite. In this paper, we consider if the least Herbrand model of the limit of a given Horn logic program sequence $\{\Pi_n\}$ equals the limit of the least Herbrand models of each logic program $\Pi_n$. It is proved that this property is not true in general but holds if Horn logic programs satisfy an assumption which can be syntactically checked and be satisfied by a class of Horn logic programs. Thus, under this assumption we can approach the least Herbrand model of the limit $\Pi$ by the sequence of the least Herbrand models of each finite program $\Pi_n$. We also prove that if a finite Horn logic program satisfies this assumption, then the least Herbrand model of this program is recursive. Finally, by use of the concept of stability from dynamical systems, we prove that this assumption is exactly a sufficient condition to guarantee the stability of fixed points for Horn logic programs.
[ { "created": "Thu, 8 Mar 2001 07:42:48 GMT", "version": "v1" }, { "created": "Fri, 9 Mar 2001 03:36:48 GMT", "version": "v2" }, { "created": "Thu, 7 Feb 2002 09:25:45 GMT", "version": "v3" } ]
2007-05-23
[ [ "Ma", "Shilong", "" ], [ "Sui", "Yuefei", "" ], [ "Xu", "Ke", "" ] ]
Given a sequence $\{\Pi_n\}$ of Horn logic programs, the limit $\Pi$ of $\{\Pi_n\}$ is the set of the clauses such that every clause in $\Pi$ belongs to almost every $\Pi_n$ and every clause in infinitely many $\Pi_n$'s belongs to $\Pi$ also. The limit program $\Pi$ is still Horn but may be infinite. In this paper, we consider if the least Herbrand model of the limit of a given Horn logic program sequence $\{\Pi_n\}$ equals the limit of the least Herbrand models of each logic program $\Pi_n$. It is proved that this property is not true in general but holds if Horn logic programs satisfy an assumption which can be syntactically checked and be satisfied by a class of Horn logic programs. Thus, under this assumption we can approach the least Herbrand model of the limit $\Pi$ by the sequence of the least Herbrand models of each finite program $\Pi_n$. We also prove that if a finite Horn logic program satisfies this assumption, then the least Herbrand model of this program is recursive. Finally, by use of the concept of stability from dynamical systems, we prove that this assumption is exactly a sufficient condition to guarantee the stability of fixed points for Horn logic programs.
2109.00632
Sriram Baireddy
Changye Yang, Sriram Baireddy, Enyu Cai, Melba Crawford, Edward J. Delp
Field-Based Plot Extraction Using UAV RGB Images
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Unmanned Aerial Vehicles (UAVs) have become popular for use in plant phenotyping of field based crops, such as maize and sorghum, due to their ability to acquire high resolution data over field trials. Field experiments, which may comprise thousands of plants, are planted according to experimental designs to evaluate varieties or management practices. For many types of phenotyping analysis, we examine smaller groups of plants known as "plots." In this paper, we propose a new plot extraction method that will segment a UAV image into plots. We will demonstrate that our method achieves higher plot extraction accuracy than existing approaches.
[ { "created": "Wed, 1 Sep 2021 22:04:59 GMT", "version": "v1" } ]
2021-09-03
[ [ "Yang", "Changye", "" ], [ "Baireddy", "Sriram", "" ], [ "Cai", "Enyu", "" ], [ "Crawford", "Melba", "" ], [ "Delp", "Edward J.", "" ] ]
Unmanned Aerial Vehicles (UAVs) have become popular for use in plant phenotyping of field based crops, such as maize and sorghum, due to their ability to acquire high resolution data over field trials. Field experiments, which may comprise thousands of plants, are planted according to experimental designs to evaluate varieties or management practices. For many types of phenotyping analysis, we examine smaller groups of plants known as "plots." In this paper, we propose a new plot extraction method that will segment a UAV image into plots. We will demonstrate that our method achieves higher plot extraction accuracy than existing approaches.
2309.16632
Andrei Graur
Andrei Graur, Haotian Jiang, Aaron Sidford
Sparse Submodular Function Minimization
Accepted to FOCS 2023
null
null
null
cs.DS math.OC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper we study the problem of minimizing a submodular function $f : 2^V \rightarrow \mathbb{R}$ that is guaranteed to have a $k$-sparse minimizer. We give a deterministic algorithm that computes an additive $\epsilon$-approximate minimizer of such $f$ in $\widetilde{O}(\mathsf{poly}(k) \log(|f|/\epsilon))$ parallel depth using a polynomial number of queries to an evaluation oracle of $f$, where $|f| = \max_{S \subseteq V} |f(S)|$. Further, we give a randomized algorithm that computes an exact minimizer of $f$ with high probability using $\widetilde{O}(|V| \cdot \mathsf{poly}(k))$ queries and polynomial time. When $k = \widetilde{O}(1)$, our algorithms use either nearly-constant parallel depth or a nearly-linear number of evaluation oracle queries. All previous algorithms for this problem either use $\Omega(|V|)$ parallel depth or $\Omega(|V|^2)$ queries. In contrast to state-of-the-art weakly-polynomial and strongly-polynomial time algorithms for SFM, our algorithms use first-order optimization methods, e.g., mirror descent and follow the regularized leader. We introduce what we call {\em sparse dual certificates}, which encode information on the structure of sparse minimizers, and both our parallel and sequential algorithms provide new algorithmic tools for allowing first-order optimization methods to efficiently compute them. Correspondingly, our algorithm does not invoke fast matrix multiplication or general linear system solvers and in this sense is more combinatorial than previous state-of-the-art methods.
[ { "created": "Thu, 28 Sep 2023 17:38:13 GMT", "version": "v1" }, { "created": "Sat, 6 Jul 2024 11:06:01 GMT", "version": "v2" } ]
2024-07-09
[ [ "Graur", "Andrei", "" ], [ "Jiang", "Haotian", "" ], [ "Sidford", "Aaron", "" ] ]
In this paper we study the problem of minimizing a submodular function $f : 2^V \rightarrow \mathbb{R}$ that is guaranteed to have a $k$-sparse minimizer. We give a deterministic algorithm that computes an additive $\epsilon$-approximate minimizer of such $f$ in $\widetilde{O}(\mathsf{poly}(k) \log(|f|/\epsilon))$ parallel depth using a polynomial number of queries to an evaluation oracle of $f$, where $|f| = \max_{S \subseteq V} |f(S)|$. Further, we give a randomized algorithm that computes an exact minimizer of $f$ with high probability using $\widetilde{O}(|V| \cdot \mathsf{poly}(k))$ queries and polynomial time. When $k = \widetilde{O}(1)$, our algorithms use either nearly-constant parallel depth or a nearly-linear number of evaluation oracle queries. All previous algorithms for this problem either use $\Omega(|V|)$ parallel depth or $\Omega(|V|^2)$ queries. In contrast to state-of-the-art weakly-polynomial and strongly-polynomial time algorithms for SFM, our algorithms use first-order optimization methods, e.g., mirror descent and follow the regularized leader. We introduce what we call {\em sparse dual certificates}, which encode information on the structure of sparse minimizers, and both our parallel and sequential algorithms provide new algorithmic tools for allowing first-order optimization methods to efficiently compute them. Correspondingly, our algorithm does not invoke fast matrix multiplication or general linear system solvers and in this sense is more combinatorial than previous state-of-the-art methods.
2407.12254
Ching Chang
Ting-Yun Ou, Ching Chang, Wen-Chih Peng
COKE: Causal Discovery with Chronological Order and Expert Knowledge in High Proportion of Missing Manufacturing Data
This paper has been accepted by the ACM International Conference on Information and Knowledge Management (CIKM) 2024
null
null
null
cs.LG stat.ME
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Understanding causal relationships between machines is crucial for fault diagnosis and optimization in manufacturing processes. Real-world datasets frequently exhibit up to 90% missing data and high dimensionality from hundreds of sensors. These datasets also include domain-specific expert knowledge and chronological order information, reflecting the recording order across different machines, which is pivotal for discerning causal relationships within the manufacturing data. However, previous methods for handling missing data in scenarios akin to real-world conditions have not been able to effectively utilize expert knowledge. Conversely, prior methods that can incorporate expert knowledge struggle with datasets that exhibit missing values. Therefore, we propose COKE to construct causal graphs in manufacturing datasets by leveraging expert knowledge and chronological order among sensors without imputing missing data. Utilizing the characteristics of the recipe, we maximize the use of samples with missing values, derive embeddings from intersections with an initial graph that incorporates expert knowledge and chronological order, and create a sensor ordering graph. The graph-generating process has been optimized by an actor-critic architecture to obtain a final graph that has a maximum reward. Experimental evaluations in diverse settings of sensor quantities and missing proportions demonstrate that our approach compared with the benchmark methods shows an average improvement of 39.9% in the F1-score. Moreover, the F1-score improvement can reach 62.6% when considering the configuration similar to real-world datasets, and 85.0% in real-world semiconductor datasets. The source code is available at https://github.com/OuTingYun/COKE.
[ { "created": "Wed, 17 Jul 2024 01:51:27 GMT", "version": "v1" }, { "created": "Thu, 1 Aug 2024 01:21:57 GMT", "version": "v2" } ]
2024-08-02
[ [ "Ou", "Ting-Yun", "" ], [ "Chang", "Ching", "" ], [ "Peng", "Wen-Chih", "" ] ]
Understanding causal relationships between machines is crucial for fault diagnosis and optimization in manufacturing processes. Real-world datasets frequently exhibit up to 90% missing data and high dimensionality from hundreds of sensors. These datasets also include domain-specific expert knowledge and chronological order information, reflecting the recording order across different machines, which is pivotal for discerning causal relationships within the manufacturing data. However, previous methods for handling missing data in scenarios akin to real-world conditions have not been able to effectively utilize expert knowledge. Conversely, prior methods that can incorporate expert knowledge struggle with datasets that exhibit missing values. Therefore, we propose COKE to construct causal graphs in manufacturing datasets by leveraging expert knowledge and chronological order among sensors without imputing missing data. Utilizing the characteristics of the recipe, we maximize the use of samples with missing values, derive embeddings from intersections with an initial graph that incorporates expert knowledge and chronological order, and create a sensor ordering graph. The graph-generating process has been optimized by an actor-critic architecture to obtain a final graph that has a maximum reward. Experimental evaluations in diverse settings of sensor quantities and missing proportions demonstrate that our approach compared with the benchmark methods shows an average improvement of 39.9% in the F1-score. Moreover, the F1-score improvement can reach 62.6% when considering the configuration similar to real-world datasets, and 85.0% in real-world semiconductor datasets. The source code is available at https://github.com/OuTingYun/COKE.
1311.1163
Zuoqiang Shi
Thomas Y. Hou, Zuoqiang Shi
Sparse Time-Frequency decomposition by dictionary learning
null
null
null
null
cs.IT math.IT
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper, we propose a time-frequency analysis method to obtain instantaneous frequencies and the corresponding decomposition by solving an optimization problem. In this optimization problem, the basis to decompose the signal is not known. Instead, it is adapted to the signal and is determined as part of the optimization problem. In this sense, this optimization problem can be seen as a dictionary learning problem. This dictionary learning problem is solved by using the Augmented Lagrangian Multiplier method (ALM) iteratively. We further accelerate the convergence of the ALM method in each iteration by using the fast wavelet transform. We apply our method to decompose several signals, including signals with poor scale separation, signals with outliers and polluted by noise and a real signal. The results show that this method can give accurate recovery of both the instantaneous frequencies and the intrinsic mode functions.
[ { "created": "Thu, 5 Sep 2013 12:55:41 GMT", "version": "v1" }, { "created": "Sat, 11 Oct 2014 08:47:30 GMT", "version": "v2" } ]
2014-10-14
[ [ "Hou", "Thomas Y.", "" ], [ "Shi", "Zuoqiang", "" ] ]
In this paper, we propose a time-frequency analysis method to obtain instantaneous frequencies and the corresponding decomposition by solving an optimization problem. In this optimization problem, the basis to decompose the signal is not known. Instead, it is adapted to the signal and is determined as part of the optimization problem. In this sense, this optimization problem can be seen as a dictionary learning problem. This dictionary learning problem is solved by using the Augmented Lagrangian Multiplier method (ALM) iteratively. We further accelerate the convergence of the ALM method in each iteration by using the fast wavelet transform. We apply our method to decompose several signals, including signals with poor scale separation, signals with outliers and polluted by noise and a real signal. The results show that this method can give accurate recovery of both the instantaneous frequencies and the intrinsic mode functions.
1711.09728
Nazmus Saquib
Md. Naimul Hoque, Rawshan E Fatima, Manash Kumar Mandal, Nazmus Saquib
Evaluating gender portrayal in Bangladeshi TV
Presented at NIPS 2017 Workshop on Machine Learning for the Developing World. Corresponding author: Nazmus Saquib
null
null
null
cs.CY stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Computer Vision and machine learning methods were previously used to reveal screen presence of genders in TV and movies. In this work, using head pose, gender detection, and skin color estimation techniques, we demonstrate that the gender disparity in TV in a South Asian country such as Bangladesh exhibits unique characteristics and is sometimes counter-intuitive to popular perception. We demonstrate a noticeable discrepancy in female screen presence in Bangladeshi TV advertisements and political talk shows. Further, contrary to popular hypotheses, we demonstrate that lighter-toned skin colors are less prevalent than darker complexions, and additionally, quantifiable body language markers do not provide conclusive insights about gender dynamics. Overall, these gender portrayal parameters reveal the different layers of onscreen gender politics and can help direct incentives to address existing disparities in a nuanced and targeted manner.
[ { "created": "Tue, 14 Nov 2017 17:54:13 GMT", "version": "v1" } ]
2017-11-29
[ [ "Hoque", "Md. Naimul", "" ], [ "Fatima", "Rawshan E", "" ], [ "Mandal", "Manash Kumar", "" ], [ "Saquib", "Nazmus", "" ] ]
Computer Vision and machine learning methods were previously used to reveal screen presence of genders in TV and movies. In this work, using head pose, gender detection, and skin color estimation techniques, we demonstrate that the gender disparity in TV in a South Asian country such as Bangladesh exhibits unique characteristics and is sometimes counter-intuitive to popular perception. We demonstrate a noticeable discrepancy in female screen presence in Bangladeshi TV advertisements and political talk shows. Further, contrary to popular hypotheses, we demonstrate that lighter-toned skin colors are less prevalent than darker complexions, and additionally, quantifiable body language markers do not provide conclusive insights about gender dynamics. Overall, these gender portrayal parameters reveal the different layers of onscreen gender politics and can help direct incentives to address existing disparities in a nuanced and targeted manner.
2207.07629
Zhiruo Zhou
Zhiruo Zhou, Hongyu Fu, Suya You, C.-C. Jay Kuo
GUSOT: Green and Unsupervised Single Object Tracking for Long Video Sequences
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Supervised and unsupervised deep trackers that rely on deep learning technologies are popular in recent years. Yet, they demand high computational complexity and a high memory cost. A green unsupervised single-object tracker, called GUSOT, that aims at object tracking for long videos under a resource-constrained environment is proposed in this work. Built upon a baseline tracker, UHP-SOT++, which works well for short-term tracking, GUSOT contains two additional new modules: 1) lost object recovery, and 2) color-saliency-based shape proposal. They help resolve the tracking loss problem and offer a more flexible object proposal, respectively. Thus, they enable GUSOT to achieve higher tracking accuracy in the long run. We conduct experiments on the large-scale dataset LaSOT with long video sequences, and show that GUSOT offers a lightweight high-performance tracking solution that finds applications in mobile and edge computing platforms.
[ { "created": "Fri, 15 Jul 2022 17:42:49 GMT", "version": "v1" } ]
2022-07-18
[ [ "Zhou", "Zhiruo", "" ], [ "Fu", "Hongyu", "" ], [ "You", "Suya", "" ], [ "Kuo", "C. -C. Jay", "" ] ]
Supervised and unsupervised deep trackers that rely on deep learning technologies are popular in recent years. Yet, they demand high computational complexity and a high memory cost. A green unsupervised single-object tracker, called GUSOT, that aims at object tracking for long videos under a resource-constrained environment is proposed in this work. Built upon a baseline tracker, UHP-SOT++, which works well for short-term tracking, GUSOT contains two additional new modules: 1) lost object recovery, and 2) color-saliency-based shape proposal. They help resolve the tracking loss problem and offer a more flexible object proposal, respectively. Thus, they enable GUSOT to achieve higher tracking accuracy in the long run. We conduct experiments on the large-scale dataset LaSOT with long video sequences, and show that GUSOT offers a lightweight high-performance tracking solution that finds applications in mobile and edge computing platforms.
2404.14739
Utkarsh Gupta
Utkarsh Gupta, Emmanouil Nikolakakis, Moritz Zaiss, Razvan Marinescu
BMapEst: Estimation of Brain Tissue Probability Maps using a Differentiable MRI Simulator
null
null
null
null
cs.CV
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Reconstructing digital brain phantoms in the form of voxel-based, multi-channeled tissue probability maps for individual subjects is essential for capturing brain anatomical variability, understanding neurological diseases, as well as for testing image processing methods. We demonstrate the first framework that estimates brain tissue probability maps (Grey Matter - GM, White Matter - WM, and Cerebrospinal fluid - CSF) with the help of a Physics-based differentiable MRI simulator that models the magnetization signal at each voxel in the volume. Given an observed $T_1$/$T_2$-weighted MRI scan, the corresponding clinical MRI sequence, and the MRI differentiable simulator, we estimate the simulator's input probability maps by back-propagating the L2 loss between the simulator's output and the $T_1$/$T_2$-weighted scan. This approach has the significant advantage of not relying on any training data and instead uses the strong inductive bias of the MRI simulator. We tested the model on 20 scans from the BrainWeb database and demonstrated a highly accurate reconstruction of GM, WM, and CSF. Our source code is available online: https://github.com/BioMedAI-UCSC/BMapEst.
[ { "created": "Tue, 23 Apr 2024 04:45:23 GMT", "version": "v1" }, { "created": "Sun, 30 Jun 2024 04:00:09 GMT", "version": "v2" } ]
2024-07-02
[ [ "Gupta", "Utkarsh", "" ], [ "Nikolakakis", "Emmanouil", "" ], [ "Zaiss", "Moritz", "" ], [ "Marinescu", "Razvan", "" ] ]
Reconstructing digital brain phantoms in the form of voxel-based, multi-channeled tissue probability maps for individual subjects is essential for capturing brain anatomical variability, understanding neurological diseases, as well as for testing image processing methods. We demonstrate the first framework that estimates brain tissue probability maps (Grey Matter - GM, White Matter - WM, and Cerebrospinal fluid - CSF) with the help of a Physics-based differentiable MRI simulator that models the magnetization signal at each voxel in the volume. Given an observed $T_1$/$T_2$-weighted MRI scan, the corresponding clinical MRI sequence, and the MRI differentiable simulator, we estimate the simulator's input probability maps by back-propagating the L2 loss between the simulator's output and the $T_1$/$T_2$-weighted scan. This approach has the significant advantage of not relying on any training data and instead uses the strong inductive bias of the MRI simulator. We tested the model on 20 scans from the BrainWeb database and demonstrated a highly accurate reconstruction of GM, WM, and CSF. Our source code is available online: https://github.com/BioMedAI-UCSC/BMapEst.
2011.10659
Mathilde Fekom
Mathilde Fekom and Argyris Kalogeratos
Efficient stream-based Max-Min diversification with minimal failure rate
null
null
null
null
cs.DS
http://creativecommons.org/licenses/by/4.0/
The stream-based Max-Min diversification problem concerns the task of selecting a limited number of diverse instances from a data stream. The nature of the problem demands immediate and irrevocable decisions. The set-wise diversity to be maximized is the minimum distance among any pair of the selected instances. Standard algorithmic approaches for sequential selection disregard the possibility of selection failures, which is the situation where the last instances of the stream are picked by default to prevent having an incomplete selection. This defect can be catastrophic for the Max-Min diversification objective. In this paper we present the Failure Rate Minimization (FRM) algorithm that allows the selection of a set of disparate instances while reducing significantly the probability of having failures. This is achieved by means of both analytical and empirical techniques. FRM is put in comparison with relevant algorithms from the literature through simulations on real datasets, where we demonstrate its efficiency and low time complexity.
[ { "created": "Tue, 17 Nov 2020 14:20:16 GMT", "version": "v1" } ]
2020-11-24
[ [ "Fekom", "Mathilde", "" ], [ "Kalogeratos", "Argyris", "" ] ]
The stream-based Max-Min diversification problem concerns the task of selecting a limited number of diverse instances from a data stream. The nature of the problem demands immediate and irrevocable decisions. The set-wise diversity to be maximized is the minimum distance among any pair of the selected instances. Standard algorithmic approaches for sequential selection disregard the possibility of selection failures, which is the situation where the last instances of the stream are picked by default to prevent having an incomplete selection. This defect can be catastrophic for the Max-Min diversification objective. In this paper we present the Failure Rate Minimization (FRM) algorithm that allows the selection of a set of disparate instances while reducing significantly the probability of having failures. This is achieved by means of both analytical and empirical techniques. FRM is put in comparison with relevant algorithms from the literature through simulations on real datasets, where we demonstrate its efficiency and low time complexity.
2209.10477
Yan Liu
Yan Liu, Maria Laricheva, Chiyu Zhang, Patrick Boutet, Guanyu Chen, Terence Tracey, Giuseppe Carenini, Richard Young
Transition to Adulthood for Young People with Intellectual or Developmental Disabilities: Emotion Detection and Topic Modeling
Conference proceedings of 2022 SBP-BRiMS
In: Thomson, R., Dancy, C., Pyke, A. (eds) SBP-BRiMS 2022. Lecture Notes in Computer Science, vol 13558. Springer, Cham (2022)
10.1007/978-3-031-17114-7_21
null
cs.CL stat.AP stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Transition to Adulthood is an essential life stage for many families. The prior research has shown that young people with intellectual or development disabil-ities (IDD) have more challenges than their peers. This study is to explore how to use natural language processing (NLP) methods, especially unsupervised machine learning, to assist psychologists to analyze emotions and sentiments and to use topic modeling to identify common issues and challenges that young people with IDD and their families have. Additionally, the results were compared to those obtained from young people without IDD who were in tran-sition to adulthood. The findings showed that NLP methods can be very useful for psychologists to analyze emotions, conduct cross-case analysis, and sum-marize key topics from conversational data. Our Python code is available at https://github.com/mlaricheva/emotion_topic_modeling.
[ { "created": "Wed, 21 Sep 2022 16:23:45 GMT", "version": "v1" } ]
2022-09-22
[ [ "Liu", "Yan", "" ], [ "Laricheva", "Maria", "" ], [ "Zhang", "Chiyu", "" ], [ "Boutet", "Patrick", "" ], [ "Chen", "Guanyu", "" ], [ "Tracey", "Terence", "" ], [ "Carenini", "Giuseppe", "" ], [ "Young", "Richard", "" ] ]
Transition to Adulthood is an essential life stage for many families. The prior research has shown that young people with intellectual or development disabil-ities (IDD) have more challenges than their peers. This study is to explore how to use natural language processing (NLP) methods, especially unsupervised machine learning, to assist psychologists to analyze emotions and sentiments and to use topic modeling to identify common issues and challenges that young people with IDD and their families have. Additionally, the results were compared to those obtained from young people without IDD who were in tran-sition to adulthood. The findings showed that NLP methods can be very useful for psychologists to analyze emotions, conduct cross-case analysis, and sum-marize key topics from conversational data. Our Python code is available at https://github.com/mlaricheva/emotion_topic_modeling.
2303.05740
Furan Xie
Bing Liu, Furan Xie and Li Chai
A Novel Bilateral Energy Trading Mechanism for Electricity Markets with Numerous Prosumers
null
null
null
null
cs.DC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
With the rapid development of distributed energy resources, increasing number of residential and commercial users have been switched from pure electricity consumers to prosumers that can both consume and produce energy. To properly manage these emerging prosumers, a peer-to-peer electricity market has been explored and extensively studied. In such an electricity market, each prosumer trades energy directly with other prosumers, posing a serious challenge to the scalability of the market. Therefore, a bilateral energy trading mechanism with good scalability is proposed for electricity markets with numerous prosumers in this paper. First, the multi-bilateral economic dispatch problem that maximizes the social welfare is formulated, taking into account product differentiation and network constraints. Then, an energy trading mechanism is devised to improve the scalability from two aspects: (i) an accelerated distributed clearing algorithm with less exchanged information and faster convergence rate. (ii) a novel selection strategy to reduce the amount of computation and communication per prosumer. Finally, the convergence proof of the proposed accelerated algorithm is given, and the proposed selection strategy is illustrated through a Monte Carlo simulation experiment.
[ { "created": "Fri, 10 Mar 2023 06:49:53 GMT", "version": "v1" }, { "created": "Tue, 12 Sep 2023 12:39:52 GMT", "version": "v2" }, { "created": "Thu, 14 Sep 2023 03:36:37 GMT", "version": "v3" } ]
2023-09-15
[ [ "Liu", "Bing", "" ], [ "Xie", "Furan", "" ], [ "Chai", "Li", "" ] ]
With the rapid development of distributed energy resources, increasing number of residential and commercial users have been switched from pure electricity consumers to prosumers that can both consume and produce energy. To properly manage these emerging prosumers, a peer-to-peer electricity market has been explored and extensively studied. In such an electricity market, each prosumer trades energy directly with other prosumers, posing a serious challenge to the scalability of the market. Therefore, a bilateral energy trading mechanism with good scalability is proposed for electricity markets with numerous prosumers in this paper. First, the multi-bilateral economic dispatch problem that maximizes the social welfare is formulated, taking into account product differentiation and network constraints. Then, an energy trading mechanism is devised to improve the scalability from two aspects: (i) an accelerated distributed clearing algorithm with less exchanged information and faster convergence rate. (ii) a novel selection strategy to reduce the amount of computation and communication per prosumer. Finally, the convergence proof of the proposed accelerated algorithm is given, and the proposed selection strategy is illustrated through a Monte Carlo simulation experiment.
2002.03586
Thomas Sturm
Hamid Rahkooy and Thomas Sturm
First-Order Tests for Toricity
null
Proc. CASC 2020, LNCS 12291, pp.510-527, Springer 2020
10.1007/978-3-030-60026-6_30
null
cs.SC q-bio.MN
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Motivated by problems arising with the symbolic analysis of steady state ideals in Chemical Reaction Network Theory, we consider the problem of testing whether the points in a complex or real variety with non-zero coordinates form a coset of a multiplicative group. That property corresponds to Shifted Toricity, a recent generalization of toricity of the corresponding polynomial ideal. The key idea is to take a geometric view on varieties rather than an algebraic view on ideals. Recently, corresponding coset tests have been proposed for complex and for real varieties. The former combine numerous techniques from commutative algorithmic algebra with Gr\"obner bases as the central algorithmic tool. The latter are based on interpreted first-order logic in real closed fields with real quantifier elimination techniques on the algorithmic side. Here we take a new logic approach to both theories, complex and real, and beyond. Besides alternative algorithms, our approach provides a unified view on theories of fields and helps to understand the relevance and interconnection of the rich existing literature in the area, which has been focusing on complex numbers, while from a scientific point of view the (positive) real numbers are clearly the relevant domain in chemical reaction network theory. We apply prototypical implementations of our new approach to a set of 129 models from the BioModels repository.
[ { "created": "Mon, 10 Feb 2020 07:56:07 GMT", "version": "v1" } ]
2020-10-22
[ [ "Rahkooy", "Hamid", "" ], [ "Sturm", "Thomas", "" ] ]
Motivated by problems arising with the symbolic analysis of steady state ideals in Chemical Reaction Network Theory, we consider the problem of testing whether the points in a complex or real variety with non-zero coordinates form a coset of a multiplicative group. That property corresponds to Shifted Toricity, a recent generalization of toricity of the corresponding polynomial ideal. The key idea is to take a geometric view on varieties rather than an algebraic view on ideals. Recently, corresponding coset tests have been proposed for complex and for real varieties. The former combine numerous techniques from commutative algorithmic algebra with Gr\"obner bases as the central algorithmic tool. The latter are based on interpreted first-order logic in real closed fields with real quantifier elimination techniques on the algorithmic side. Here we take a new logic approach to both theories, complex and real, and beyond. Besides alternative algorithms, our approach provides a unified view on theories of fields and helps to understand the relevance and interconnection of the rich existing literature in the area, which has been focusing on complex numbers, while from a scientific point of view the (positive) real numbers are clearly the relevant domain in chemical reaction network theory. We apply prototypical implementations of our new approach to a set of 129 models from the BioModels repository.
2006.01780
Rahat Yeasin Emon
Rahat Yeasin Emon
A Novel Nudity Detection Algorithm for Web and Mobile Application Development
5 pages
null
null
null
cs.CV cs.MM
http://creativecommons.org/licenses/by/4.0/
In our current web and mobile application development runtime nude image content detection is very important. This paper presents a runtime nudity detection method for web and mobile application development. We use two parameters to detect the nude content of an image. One is the number of skin pixels another is face region. A skin color model based on RGB, HSV color spaces are used to detect skin pixels in an image. Google vision api is used to detect the face region. By the percentage of skin regions and face regions an image is identified nude or not. The success of this algorithm exists in detecting skin regions and face regions. The skin detection algorithm can detect skin 95% accurately with a low false-positive rate and the google vision api for web and mobile applications can detect face 99% accurately with less than 1 second time. From the experimental analysis, we have seen that the proposed algorithm can detect 95% percent accurately the nudity of an image.
[ { "created": "Tue, 2 Jun 2020 17:00:47 GMT", "version": "v1" }, { "created": "Sun, 28 Jun 2020 15:29:09 GMT", "version": "v2" } ]
2020-06-30
[ [ "Emon", "Rahat Yeasin", "" ] ]
In our current web and mobile application development runtime nude image content detection is very important. This paper presents a runtime nudity detection method for web and mobile application development. We use two parameters to detect the nude content of an image. One is the number of skin pixels another is face region. A skin color model based on RGB, HSV color spaces are used to detect skin pixels in an image. Google vision api is used to detect the face region. By the percentage of skin regions and face regions an image is identified nude or not. The success of this algorithm exists in detecting skin regions and face regions. The skin detection algorithm can detect skin 95% accurately with a low false-positive rate and the google vision api for web and mobile applications can detect face 99% accurately with less than 1 second time. From the experimental analysis, we have seen that the proposed algorithm can detect 95% percent accurately the nudity of an image.
2008.08675
Anders Johan Andreassen
Anders Andreassen, Ethan Dyer
Asymptotics of Wide Convolutional Neural Networks
23 pages, 12 figures
null
null
null
cs.LG hep-th stat.ML
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Wide neural networks have proven to be a rich class of architectures for both theory and practice. Motivated by the observation that finite width convolutional networks appear to outperform infinite width networks, we study scaling laws for wide CNNs and networks with skip connections. Following the approach of (Dyer & Gur-Ari, 2019), we present a simple diagrammatic recipe to derive the asymptotic width dependence for many quantities of interest. These scaling relationships provide a solvable description for the training dynamics of wide convolutional networks. We test these relations across a broad range of architectures. In particular, we find that the difference in performance between finite and infinite width models vanishes at a definite rate with respect to model width. Nonetheless, this relation is consistent with finite width models generalizing either better or worse than their infinite width counterparts, and we provide examples where the relative performance depends on the optimization details.
[ { "created": "Wed, 19 Aug 2020 21:22:19 GMT", "version": "v1" } ]
2020-08-21
[ [ "Andreassen", "Anders", "" ], [ "Dyer", "Ethan", "" ] ]
Wide neural networks have proven to be a rich class of architectures for both theory and practice. Motivated by the observation that finite width convolutional networks appear to outperform infinite width networks, we study scaling laws for wide CNNs and networks with skip connections. Following the approach of (Dyer & Gur-Ari, 2019), we present a simple diagrammatic recipe to derive the asymptotic width dependence for many quantities of interest. These scaling relationships provide a solvable description for the training dynamics of wide convolutional networks. We test these relations across a broad range of architectures. In particular, we find that the difference in performance between finite and infinite width models vanishes at a definite rate with respect to model width. Nonetheless, this relation is consistent with finite width models generalizing either better or worse than their infinite width counterparts, and we provide examples where the relative performance depends on the optimization details.
1702.03186
Matthieu Guillot
Matthieu Guillot and Gautier Stauffer
The Stochastic Shortest Path Problem : A polyhedral combinatorics perspective
null
null
null
null
cs.DM math.OC
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
In this paper, we give a new framework for the stochastic shortest path problem in finite state and action spaces. Our framework generalizes both the frameworks proposed by Bertsekas and Tsitsikli and by Bertsekas and Yu. We prove that the problem is well-defined and (weakly) polynomial when (i) there is a way to reach the target state from any initial state and (ii) there is no transition cycle of negative costs (a generalization of negative cost cycles). These assumptions generalize the standard assumptions for the deterministic shortest path problem and our framework encapsulates the latter problem (in contrast with prior works). In this new setting, we can show that (a) one can restrict to deterministic and stationary policies, (b) the problem is still (weakly) polynomial through linear programming, (c) Value Iteration and Policy Iteration converge, and (d) we can extend Dijkstra's algorithm.
[ { "created": "Fri, 10 Feb 2017 14:36:32 GMT", "version": "v1" } ]
2017-02-13
[ [ "Guillot", "Matthieu", "" ], [ "Stauffer", "Gautier", "" ] ]
In this paper, we give a new framework for the stochastic shortest path problem in finite state and action spaces. Our framework generalizes both the frameworks proposed by Bertsekas and Tsitsikli and by Bertsekas and Yu. We prove that the problem is well-defined and (weakly) polynomial when (i) there is a way to reach the target state from any initial state and (ii) there is no transition cycle of negative costs (a generalization of negative cost cycles). These assumptions generalize the standard assumptions for the deterministic shortest path problem and our framework encapsulates the latter problem (in contrast with prior works). In this new setting, we can show that (a) one can restrict to deterministic and stationary policies, (b) the problem is still (weakly) polynomial through linear programming, (c) Value Iteration and Policy Iteration converge, and (d) we can extend Dijkstra's algorithm.