JohnVitz/NLP_Assignment_3_part_3 · Datasets at Hugging Face

1802.00510

Adaptive Memory Networks

We present Adaptive Memory Networks (AMN) that processes input-question pairs to dynamically construct a network architecture optimized for lower inference times for Question Answering (QA) tasks. AMN processes the input story to extract entities and stores them in memory banks. Starting from a single bank, as the number of input entities increases, AMN learns to create new banks as the entropy in a single bank becomes too high. Hence, after processing an input-question(s) pair, the resulting network represents a hierarchical structure where entities are stored in different banks, distanced by question relevance. At inference, one or few banks are used, creating a tradeoff between accuracy and performance. AMN is enabled by dynamic networks that allow input dependent network creation and efficiency in dynamic mini-batching as well as our novel bank controller that allows learning discrete decision making with high accuracy. In our results, we demonstrate that AMN learns to create variable depth networks depending on task complexity and reduces inference times for QA tasks.

false

true

false

true

false

89,426

2412.20620

Matrix Concentration for Random Signed Graphs and Community Recovery in the Signed Stochastic Block Model

We consider graphs where edges and their signs are added independently at random from among all pairs of nodes. We establish strong concentration inequalities for adjacency and Laplacian matrices obtained from this family of random graph models. Then, we apply our results to study graphs sampled from the signed stochastic block model. Namely, we take a two-community setting where edges within the communities have positive signs and edges between the communities have negative signs and apply a random sign perturbation with probability $0< s <1/2$. In this setting, our findings include: first, the spectral gap of the corresponding signed Laplacian matrix concentrates near $2s$ with high probability; and second, the sign of the first eigenvector of the Laplacian matrix defines a weakly consistent estimator for the balanced community detection problem, or equivalently, the $\pm 1$ synchronization problem. We supplement our theoretical contributions with experimental data obtained from the models under consideration.

false

true

false

true

false

521,287

1502.00827

On the Duality of Additivity and Tensorization

A function is said to be additive if, similar to mutual information, expands by a factor of $n$, when evaluated on $n$ i.i.d. repetitions of a source or channel. On the other hand, a function is said to satisfy the tensorization property if it remains unchanged when evaluated on i.i.d. repetitions. Additive rate regions are of fundamental importance in network information theory, serving as capacity regions or upper bounds thereof. Tensorizing measures of correlation have also found applications in distributed source and channel coding problems as well as the distribution simulation problem. Prior to our work only two measures of correlation, namely the hypercontractivity ribbon and maximal correlation (and their derivatives), were known to have the tensorization property. In this paper, we provide a general framework to obtain a region with the tensorization property from any additive rate region. We observe that hypercontractivity ribbon indeed comes from the dual of the rate region of the Gray-Wyner source coding problem, and generalize it to the multipartite case. Then we define other measures of correlation with similar properties from other source coding problems. We also present some applications of our results.

false

true

false

39,878

2409.05477

Retrofitting Temporal Graph Neural Networks with Transformer

Temporal graph neural networks (TGNNs) outperform regular GNNs by incorporating time information into graph-based operations. However, TGNNs adopt specialized models (e.g., TGN, TGAT, and APAN ) and require tailored training frameworks (e.g., TGL and ETC). In this paper, we propose TF-TGN, which uses Transformer decoder as the backbone model for TGNN to enjoy Transformer's codebase for efficient training. In particular, Transformer achieves tremendous success for language modeling, and thus the community developed high-performance kernels (e.g., flash-attention and memory-efficient attention) and efficient distributed training schemes (e.g., PyTorch FSDP, DeepSpeed, and Megatron-LM). We observe that TGNN resembles language modeling, i.e., the message aggregation operation between chronologically occurring nodes and their temporal neighbors in TGNNs can be structured as sequence modeling. Beside this similarity, we also incorporate a series of algorithm designs including suffix infilling, temporal graph attention with self-loop, and causal masking self-attention to make TF-TGN work. During training, existing systems are slow in transforming the graph topology and conducting graph sampling. As such, we propose methods to parallelize the CSR format conversion and graph sampling. We also adapt Transformer codebase to train TF-TGN efficiently with multiple GPUs. We experiment with 9 graphs and compare with 2 state-of-the-art TGNN training frameworks. The results show that TF-TGN can accelerate training by over 2.20 while providing comparable or even superior accuracy to existing SOTA TGNNs. TF-TGN is available at https://github.com/qianghuangwhu/TF-TGN.

false

true

false

486,789

1611.03765

Emulating Batteries with Deferrable Energy Demand: Fundamental Trade-offs and Scheduling Policies

We investigate the ability of a homogeneous collection of deferrable energy loads to behave as a battery; that is, to absorb and release energy in a controllable fashion up to fixed and predetermined limits on volume, charge rate and discharge rate. We derive explicit bounds on the battery capacity that can be offered, and show that there is a fundamental trade-off between the abilities of collective load to absorb and release energy at high aggregate rates. Finally, we introduce a new class of dynamic priority-driven feedback policies that balance these abilities, and characterize the batteries that they can emulate.

false

true

false

63,740

2304.09084

DRIFT: A Federated Recommender System with Implicit Feedback on the Items

Nowadays there are more and more items available online, this makes it hard for users to find items that they like. Recommender systems aim to find the item who best suits the user, using his historical interactions. Depending on the context, these interactions may be more or less sensitive and collecting them brings an important problem concerning the users' privacy. Federated systems have shown that it is possible to make accurate and efficient recommendations without storing users' personal information. However, these systems use instantaneous feedback from the user. In this report, we propose DRIFT, a federated architecture for recommender systems, using implicit feedback. Our learning model is based on a recent algorithm for recommendation with implicit feedbacks SAROS. We aim to make recommendations as precise as SAROS, without compromising the users' privacy. In this report we show that thanks to our experiments, but also thanks to a theoretical analysis on the convergence. We have shown also that the computation time has a linear complexity with respect to the number of interactions made. Finally, we have shown that our algorithm is secure, and participants in our federated system cannot guess the interactions made by the user, except DOs that have the item involved in the interaction.

false

true

false

358,929

2305.10839

A Lexical-aware Non-autoregressive Transformer-based ASR Model

Non-autoregressive automatic speech recognition (ASR) has become a mainstream of ASR modeling because of its fast decoding speed and satisfactory result. To further boost the performance, relaxing the conditional independence assumption and cascading large-scaled pre-trained models are two active research directions. In addition to these strategies, we propose a lexical-aware non-autoregressive Transformer-based (LA-NAT) ASR framework, which consists of an acoustic encoder, a speech-text shared encoder, and a speech-text shared decoder. The acoustic encoder is used to process the input speech features as usual, and the speech-text shared encoder and decoder are designed to train speech and text data simultaneously. By doing so, LA-NAT aims to make the ASR model aware of lexical information, so the resulting model is expected to achieve better results by leveraging the learned linguistic knowledge. A series of experiments are conducted on the AISHELL-1, CSJ, and TEDLIUM 2 datasets. According to the experiments, the proposed LA-NAT can provide superior results than other recently proposed non-autoregressive ASR models. In addition, LA-NAT is a relatively compact model than most non-autoregressive ASR models, and it is about 58 times faster than the classic autoregressive model.

false

true

false

true

false

365,252

1905.13594

Known-plaintext attack and ciphertext-only attack for encrypted single-pixel imaging

In many previous works, a single-pixel imaging (SPI) system is constructed as an optical image encryption system. Unauthorized users are not able to reconstruct the plaintext image from the ciphertext intensity sequence without knowing the illumination pattern key. However, little cryptanalysis about encrypted SPI has been investigated in the past. In this work, we propose a known-plaintext attack scheme and a ciphertext-only attack scheme to an encrypted SPI system for the first time. The known-plaintext attack is implemented by interchanging the roles of illumination patterns and object images in the SPI model. The ciphertext-only attack is implemented based on the statistical features of single-pixel intensity values. The two schemes can crack encrypted SPI systems and successfully recover the key containing correct illumination patterns.

false

true

false

133,185

2305.18453

Conditional Diffusion Models for Semantic 3D Brain MRI Synthesis

Artificial intelligence (AI) in healthcare, especially in medical imaging, faces challenges due to data scarcity and privacy concerns. Addressing these, we introduce Med-DDPM, a diffusion model designed for 3D semantic brain MRI synthesis. This model effectively tackles data scarcity and privacy issues by integrating semantic conditioning. This involves the channel-wise concatenation of a conditioning image to the model input, enabling control in image generation. Med-DDPM demonstrates superior stability and performance compared to existing 3D brain imaging synthesis methods. It generates diverse, anatomically coherent images with high visual fidelity. In terms of dice score accuracy in the tumor segmentation task, Med-DDPM achieves 0.6207, close to the 0.6531 accuracy of real images, and outperforms baseline models. Combined with real images, it further increases segmentation accuracy to 0.6675, showing the potential of our proposed method for data augmentation. This model represents the first use of a diffusion model in 3D semantic brain MRI synthesis, producing high-quality images. Its semantic conditioning feature also shows potential for image anonymization in biomedical imaging, addressing data and privacy issues. We provide the code and model weights for Med-DDPM on our GitHub repository (https://github.com/mobaidoctor/med-ddpm/) to support reproducibility.

false

true

false

true

false

369,069

2304.02744

StyleGAN Salon: Multi-View Latent Optimization for Pose-Invariant Hairstyle Transfer

Our paper seeks to transfer the hairstyle of a reference image to an input photo for virtual hair try-on. We target a variety of challenges scenarios, such as transforming a long hairstyle with bangs to a pixie cut, which requires removing the existing hair and inferring how the forehead would look, or transferring partially visible hair from a hat-wearing person in a different pose. Past solutions leverage StyleGAN for hallucinating any missing parts and producing a seamless face-hair composite through so-called GAN inversion or projection. However, there remains a challenge in controlling the hallucinations to accurately transfer hairstyle and preserve the face shape and identity of the input. To overcome this, we propose a multi-view optimization framework that uses "two different views" of reference composites to semantically guide occluded or ambiguous regions. Our optimization shares information between two poses, which allows us to produce high fidelity and realistic results from incomplete references. Our framework produces high-quality results and outperforms prior work in a user study that consists of significantly more challenging hair transfer scenarios than previously studied. Project page: https://stylegan-salon.github.io/.

false

true

false

356,529

2304.05836

A Game-theoretic Framework for Privacy-preserving Federated Learning

In federated learning, benign participants aim to optimize a global model collaboratively. However, the risk of \textit{privacy leakage} cannot be ignored in the presence of \textit{semi-honest} adversaries. Existing research has focused either on designing protection mechanisms or on inventing attacking mechanisms. While the battle between defenders and attackers seems never-ending, we are concerned with one critical question: is it possible to prevent potential attacks in advance? To address this, we propose the first game-theoretic framework that considers both FL defenders and attackers in terms of their respective payoffs, which include computational costs, FL model utilities, and privacy leakage risks. We name this game the federated learning privacy game (FLPG), in which neither defenders nor attackers are aware of all participants' payoffs. To handle the \textit{incomplete information} inherent in this situation, we propose associating the FLPG with an \textit{oracle} that has two primary responsibilities. First, the oracle provides lower and upper bounds of the payoffs for the players. Second, the oracle acts as a correlation device, privately providing suggested actions to each player. With this novel framework, we analyze the optimal strategies of defenders and attackers. Furthermore, we derive and demonstrate conditions under which the attacker, as a rational decision-maker, should always follow the oracle's suggestion \textit{not to attack}.

false

true

false

true

false

true

false

true

357,759

2102.01255

TinyML for Ubiquitous Edge AI

TinyML is a fast-growing multidisciplinary field at the intersection of machine learning, hardware, and software, that focuses on enabling deep learning algorithms on embedded (microcontroller powered) devices operating at extremely low power range (mW range and below). TinyML addresses the challenges in designing power-efficient, compact deep neural network models, supporting software framework, and embedded hardware that will enable a wide range of customized, ubiquitous inference applications on battery-operated, resource-constrained devices. In this report, we discuss the major challenges and technological enablers that direct this field's expansion. TinyML will open the door to the new types of edge services and applications that do not rely on cloud processing but thrive on distributed edge inference and autonomous reasoning.

false

true

false

218,044

2309.07861

CiwaGAN: Articulatory information exchange

Humans encode information into sounds by controlling articulators and decode information from sounds using the auditory apparatus. This paper introduces CiwaGAN, a model of human spoken language acquisition that combines unsupervised articulatory modeling with an unsupervised model of information exchange through the auditory modality. While prior research includes unsupervised articulatory modeling and information exchange separately, our model is the first to combine the two components. The paper also proposes an improved articulatory model with more interpretable internal representations. The proposed CiwaGAN model is the most realistic approximation of human spoken language acquisition using deep learning. As such, it is useful for cognitively plausible simulations of the human speech act.

false

true

false

true

false

true

false

391,929

2304.04222

CILIATE: Towards Fairer Class-based Incremental Learning by Dataset and Training Refinement

Due to the model aging problem, Deep Neural Networks (DNNs) need updates to adjust them to new data distributions. The common practice leverages incremental learning (IL), e.g., Class-based Incremental Learning (CIL) that updates output labels, to update the model with new data and a limited number of old data. This avoids heavyweight training (from scratch) using conventional methods and saves storage space by reducing the number of old data to store. But it also leads to poor performance in fairness. In this paper, we show that CIL suffers both dataset and algorithm bias problems, and existing solutions can only partially solve the problem. We propose a novel framework, CILIATE, that fixes both dataset and algorithm bias in CIL. It features a novel differential analysis guided dataset and training refinement process that identifies unique and important samples overlooked by existing CIL and enforces the model to learn from them. Through this process, CILIATE improves the fairness of CIL by 17.03%, 22.46%, and 31.79% compared to state-of-the-art methods, iCaRL, BiC, and WA, respectively, based on our evaluation on three popular datasets and widely used ResNet models.

false

true

false

true

false

true

false

true

357,137

2005.05587

Robustness Verification for Classifier Ensembles

We give a formal verification procedure that decides whether a classifier ensemble is robust against arbitrary randomized attacks. Such attacks consist of a set of deterministic attacks and a distribution over this set. The robustness-checking problem consists of assessing, given a set of classifiers and a labelled data set, whether there exists a randomized attack that induces a certain expected loss against all classifiers. We show the NP-hardness of the problem and provide an upper bound on the number of attacks that is sufficient to form an optimal randomized attack. These results provide an effective way to reason about the robustness of a classifier ensemble. We provide SMT and MILP encodings to compute optimal randomized attacks or prove that there is no attack inducing a certain expected loss. In the latter case, the classifier ensemble is provably robust. Our prototype implementation verifies multiple neural-network ensembles trained for image-classification tasks. The experimental results using the MILP encoding are promising both in terms of scalability and the general applicability of our verification procedure.

false

true

false

176,776

2305.13031

HGFormer: Hierarchical Grouping Transformer for Domain Generalized Semantic Segmentation

Current semantic segmentation models have achieved great success under the independent and identically distributed (i.i.d.) condition. However, in real-world applications, test data might come from a different domain than training data. Therefore, it is important to improve model robustness against domain differences. This work studies semantic segmentation under the domain generalization setting, where a model is trained only on the source domain and tested on the unseen target domain. Existing works show that Vision Transformers are more robust than CNNs and show that this is related to the visual grouping property of self-attention. In this work, we propose a novel hierarchical grouping transformer (HGFormer) to explicitly group pixels to form part-level masks and then whole-level masks. The masks at different scales aim to segment out both parts and a whole of classes. HGFormer combines mask classification results at both scales for class label prediction. We assemble multiple interesting cross-domain settings by using seven public semantic segmentation datasets. Experiments show that HGFormer yields more robust semantic segmentation results than per-pixel classification methods and flat grouping transformers, and outperforms previous methods significantly. Code will be available at https://github.com/dingjiansw101/HGFormer.

false

true

false

366,305

2402.05467

Rapid Optimization for Jailbreaking LLMs via Subconscious Exploitation and Echopraxia

Large Language Models (LLMs) have become prevalent across diverse sectors, transforming human life with their extraordinary reasoning and comprehension abilities. As they find increased use in sensitive tasks, safety concerns have gained widespread attention. Extensive efforts have been dedicated to aligning LLMs with human moral principles to ensure their safe deployment. Despite their potential, recent research indicates aligned LLMs are prone to specialized jailbreaking prompts that bypass safety measures to elicit violent and harmful content. The intrinsic discrete nature and substantial scale of contemporary LLMs pose significant challenges in automatically generating diverse, efficient, and potent jailbreaking prompts, representing a continuous obstacle. In this paper, we introduce RIPPLE (Rapid Optimization via Subconscious Exploitation and Echopraxia), a novel optimization-based method inspired by two psychological concepts: subconsciousness and echopraxia, which describe the processes of the mind that occur without conscious awareness and the involuntary mimicry of actions, respectively. Evaluations across 6 open-source LLMs and 4 commercial LLM APIs show RIPPLE achieves an average Attack Success Rate of 91.5\%, outperforming five current methods by up to 47.0\% with an 8x reduction in overhead. Furthermore, it displays significant transferability and stealth, successfully evading established detection mechanisms. The code of our work is available at \url{https://github.com/SolidShen/RIPPLE_official/tree/official}

false

true

false

true

false

true

false

427,873

2408.09196

Maintainability Challenges in ML: A Systematic Literature Review

Background: As Machine Learning (ML) advances rapidly in many fields, it is being adopted by academics and businesses alike. However, ML has a number of different challenges in terms of maintenance not found in traditional software projects. Identifying what causes these maintainability challenges can help mitigate them early and continue delivering value in the long run without degrading ML performance. Aim: This study aims to identify and synthesise the maintainability challenges in different stages of the ML workflow and understand how these stages are interdependent and impact each other's maintainability. Method: Using a systematic literature review, we screened more than 13000 papers, then selected and qualitatively analysed 56 of them. Results: (i) a catalogue of maintainability challenges in different stages of Data Engineering, Model Engineering workflows and the current challenges when building ML systems are discussed; (ii) a map of 13 maintainability challenges to different interdependent stages of ML that impact the overall workflow; (iii) Provided insights to developers of ML tools and researchers. Conclusions: In this study, practitioners and organisations will learn about maintainability challenges and their impact at different stages of ML workflow. This will enable them to avoid pitfalls and help to build a maintainable ML system. The implications and challenges will also serve as a basis for future research to strengthen our understanding of the ML system's maintainability.

false

true

false

true

481,326

1408.0853

Adaptive Learning in Cartesian Product of Reproducing Kernel Hilbert Spaces

We propose a novel adaptive learning algorithm based on iterative orthogonal projections in the Cartesian product of multiple reproducing kernel Hilbert spaces (RKHSs). The task is estimating/tracking nonlinear functions which are supposed to contain multiple components such as (i) linear and nonlinear components, (ii) high- and low- frequency components etc. In this case, the use of multiple RKHSs permits a compact representation of multicomponent functions. The proposed algorithm is where two different methods of the author meet: multikernel adaptive filtering and the algorithm of hyperplane projection along affine subspace (HYPASS). In a certain particular case, the sum space of the RKHSs is isomorphic to the product space and hence the proposed algorithm can also be regarded as an iterative projection method in the sum space. The efficacy of the proposed algorithm is shown by numerical examples.

false

true

false

35,116

2002.09304

Stochastic Runge-Kutta methods and adaptive SGD-G2 stochastic gradient descent

The minimization of the loss function is of paramount importance in deep neural networks. On the other hand, many popular optimization algorithms have been shown to correspond to some evolution equation of gradient flow type. Inspired by the numerical schemes used for general evolution equations we introduce a second order stochastic Runge Kutta method and show that it yields a consistent procedure for the minimization of the loss function. In addition it can be coupled, in an adaptive framework, with a Stochastic Gradient Descent (SGD) to adjust automatically the learning rate of the SGD, without the need of any additional information on the Hessian of the loss functional. The adaptive SGD, called SGD-G2, is successfully tested on standard datasets.

false

true

false

true

165,031

2004.05671

Direction of Arrival Estimation for a Vector Sensor Using Deep Neural Networks

A vector sensor, a type of sensor array with six collocated antennas to measure all electromagnetic field components of incident waves, has been shown to be advantageous in estimating the angle of arrival and polarization of the incident sources. While angle estimation with machine learning for linear arrays has been well studied, there has not been a similar solution for the vector sensor. In this paper, we propose neural networks to determine the number of the sources and estimate the angle of arrival of each source, based on the covariance matrix extracted from received data. Also, we provide a solution for matching output angles to corresponding sources and examine the error distributions with this method. The results show that neural networks can achieve reasonably accurate estimation with up to 5 sources, especially if the field-of-view is limited.

false

true

false

172,267

2005.01115

Deep Encoder-Decoder Neural Network for Fingerprint Image Denoising and Inpainting

Fingerprint image denoising is a very important step in fingerprint identification. to improve the denoising effect of fingerprint image,we have designs a fingerprint denoising algorithm based on deep encoder-decoder network,which encoder subnet to learn the fingerprint features of noisy images.the decoder subnet reconstructs the original fingerprint image based on the features to achieve denoising, while using the dilated convolution in the network to increase the receptor field without increasing the complexity and improve the network inference speed. In addition, feature fusion at different levels of the network is achieved through the introduction of residual learning, which further restores the detailed features of the fingerprint and improves the denoising effect. Finally, the experimental results show that the algorithm enables better recovery of edge, line and curve features in fingerprint images, with better visual effects and higher peak signal-to-noise ratio (PSNR) compared to other methods.

false

true

false

175,494

1903.08352

GRIP: Generative Robust Inference and Perception for Semantic Robot Manipulation in Adversarial Environments

Recent advancements have led to a proliferation of machine learning systems used to assist humans in a wide range of tasks. However, we are still far from accurate, reliable, and resource-efficient operations of these systems. For robot perception, convolutional neural networks (CNNs) for object detection and pose estimation are recently coming into widespread use. However, neural networks are known to suffer overfitting during training process and are less robust within unseen conditions, which are especially vulnerable to adversarial scenarios. In this work, we propose Generative Robust Inference and Perception (GRIP) as a two-stage object detection and pose estimation system that aims to combine relative strengths of discriminative CNNs and generative inference methods to achieve robust estimation. Our results show that a second stage of sample-based generative inference is able to recover from false object detection by CNNs, and produce robust estimations in adversarial conditions. We demonstrate the efficacy of GRIP robustness through comparison with state-of-the-art learning-based pose estimators and pick-and-place manipulation in dark and cluttered environments.

false

true

false

true

false

124,815

1909.11289

Deep learning vessel segmentation and quantification of the foveal avascular zone using commercial and prototype OCT-A platforms

Automatic quantification of perifoveal vessel densities in optical coherence tomography angiography (OCT-A) images face challenges such as variable intra- and inter-image signal to noise ratios, projection artefacts from outer vasculature layers, and motion artefacts. This study demonstrates the utility of deep neural networks for automatic quantification of foveal avascular zone (FAZ) parameters and perifoveal vessel density of OCT-A images in healthy and diabetic eyes. OCT-A images of the foveal region were acquired using three OCT-A systems: a 1060nm Swept Source (SS)-OCT prototype, RTVue XR Avanti (Optovue Inc., Fremont, CA), and the ZEISS Angioplex (Carl Zeiss Meditec, Dublin, CA). Automated segmentation was then performed using a deep neural network. Four FAZ morphometric parameters (area, min/max diameter, and eccentricity) and perifoveal vessel density were used as outcome measures. The accuracy, sensitivity and specificity of the DNN vessel segmentations were comparable across all three device platforms. No significant difference between the means of the measurements from automated and manual segmentations were found for any of the outcome measures on any system. The intraclass correlation coefficient (ICC) was also good (> 0.51) for all measurements. Automated deep learning vessel segmentation of OCT-A may be suitable for both commercial and research purposes for better quantification of the retinal circulation.

false

true

false

true

false

146,771

2405.19928

BAN: Detecting Backdoors Activated by Adversarial Neuron Noise

Backdoor attacks on deep learning represent a recent threat that has gained significant attention in the research community. Backdoor defenses are mainly based on backdoor inversion, which has been shown to be generic, model-agnostic, and applicable to practical threat scenarios. State-of-the-art backdoor inversion recovers a mask in the feature space to locate prominent backdoor features, where benign and backdoor features can be disentangled. However, it suffers from high computational overhead, and we also find that it overly relies on prominent backdoor features that are highly distinguishable from benign features. To tackle these shortcomings, this paper improves backdoor feature inversion for backdoor detection by incorporating extra neuron activation information. In particular, we adversarially increase the loss of backdoored models with respect to weights to activate the backdoor effect, based on which we can easily differentiate backdoored and clean models. Experimental results demonstrate our defense, BAN, is 1.37$\times$ (on CIFAR-10) and 5.11$\times$ (on ImageNet200) more efficient with an average 9.99\% higher detect success rate than the state-of-the-art defense BTI-DBF. Our code and trained models are publicly available at~\url{https://github.com/xiaoyunxxy/ban}.

false

true

false

true

false

459,126

2406.00432

Localize, Understand, Collaborate: Semantic-Aware Dragging via Intention Reasoner

Flexible and accurate drag-based editing is a challenging task that has recently garnered significant attention. Current methods typically model this problem as automatically learning "how to drag" through point dragging and often produce one deterministic estimation, which presents two key limitations: 1) Overlooking the inherently ill-posed nature of drag-based editing, where multiple results may correspond to a given input, as illustrated in Fig.1; 2) Ignoring the constraint of image quality, which may lead to unexpected distortion. To alleviate this, we propose LucidDrag, which shifts the focus from "how to drag" to "what-then-how" paradigm. LucidDrag comprises an intention reasoner and a collaborative guidance sampling mechanism. The former infers several optimal editing strategies, identifying what content and what semantic direction to be edited. Based on the former, the latter addresses "how to drag" by collaboratively integrating existing editing guidance with the newly proposed semantic guidance and quality guidance. Specifically, semantic guidance is derived by establishing a semantic editing direction based on reasoned intentions, while quality guidance is achieved through classifier guidance using an image fidelity discriminator. Both qualitative and quantitative comparisons demonstrate the superiority of LucidDrag over previous methods.

false

true

false

459,851

2408.14916

Towards Real-world Event-guided Low-light Video Enhancement and Deblurring

In low-light conditions, capturing videos with frame-based cameras often requires long exposure times, resulting in motion blur and reduced visibility. While frame-based motion deblurring and low-light enhancement have been studied, they still pose significant challenges. Event cameras have emerged as a promising solution for improving image quality in low-light environments and addressing motion blur. They provide two key advantages: capturing scene details well even in low light due to their high dynamic range, and effectively capturing motion information during long exposures due to their high temporal resolution. Despite efforts to tackle low-light enhancement and motion deblurring using event cameras separately, previous work has not addressed both simultaneously. To explore the joint task, we first establish real-world datasets for event-guided low-light enhancement and deblurring using a hybrid camera system based on beam splitters. Subsequently, we introduce an end-to-end framework to effectively handle these tasks. Our framework incorporates a module to efficiently leverage temporal information from events and frames. Furthermore, we propose a module to utilize cross-modal feature information to employ a low-pass filter for noise suppression while enhancing the main structural information. Our proposed method significantly outperforms existing approaches in addressing the joint task. Our project pages are available at https://github.com/intelpro/ELEDNet.

false

true

false

483,733

2407.15687

SoftCVI: Contrastive variational inference with self-generated soft labels

Estimating a distribution given access to its unnormalized density is pivotal in Bayesian inference, where the posterior is generally known only up to an unknown normalizing constant. Variational inference and Markov chain Monte Carlo methods are the predominant tools for this task; however, both are often challenging to apply reliably, particularly when the posterior has complex geometry. Here, we introduce Soft Contrastive Variational Inference (SoftCVI), which allows a family of variational objectives to be derived through a contrastive estimation framework. The approach parameterizes a classifier in terms of a variational distribution, reframing the inference task as a contrastive estimation problem aiming to identify a single true posterior sample among a set of samples. Despite this framing, we do not require positive or negative samples, but rather learn by sampling the variational distribution and computing ground truth soft classification labels from the unnormalized posterior itself. The objectives have zero variance gradient when the variational approximation is exact, without the need for specialized gradient estimators. We empirically investigate the performance on a variety of Bayesian inference tasks, using both simple (e.g. normal) and expressive (normalizing flow) variational distributions. We find that SoftCVI can be used to form objectives which are stable to train and mass-covering, frequently outperforming inference with other variational approaches.

false

true

false

475,283

1906.05916

Link Dimension and Exact Construction of a Graph

Minimum resolution set and associated metric dimension provide the basis for unique and systematic labeling of nodes of a graph using distances to a set of landmarks. Such a distance vector set, however, may not be unique to the graph and does not allow for its exact construction. The concept of construction set is presented, which facilitates the unique representation of nodes and the graph as well as its exact construction. Link dimension is the minimum number of landmarks in a construction set. Results presented include necessary conditions for a set of landmarks to be a construction set, bounds for link dimension, and guidelines for transforming a resolution set to a construction set.

false

true

false

true

135,160

2203.12826

HM: Hybrid Masking for Few-Shot Segmentation

We study few-shot semantic segmentation that aims to segment a target object from a query image when provided with a few annotated support images of the target class. Several recent methods resort to a feature masking (FM) technique to discard irrelevant feature activations which eventually facilitates the reliable prediction of segmentation mask. A fundamental limitation of FM is the inability to preserve the fine-grained spatial details that affect the accuracy of segmentation mask, especially for small target objects. In this paper, we develop a simple, effective, and efficient approach to enhance feature masking (FM). We dub the enhanced FM as hybrid masking (HM). Specifically, we compensate for the loss of fine-grained spatial details in FM technique by investigating and leveraging a complementary basic input masking method. Experiments have been conducted on three publicly available benchmarks with strong few-shot segmentation (FSS) baselines. We empirically show improved performance against the current state-of-the-art methods by visible margins across different benchmarks. Our code and trained models are available at: https://github.com/moonsh/HM-Hybrid-Masking

false

true

false

287,404

1605.07774

Generalized Mirror Descents in Congestion Games

Different types of dynamics have been studied in repeated game play, and one of them which has received much attention recently consists of those based on "no-regret" algorithms from the area of machine learning. It is known that dynamics based on generic no-regret algorithms may not converge to Nash equilibria in general, but to a larger set of outcomes, namely coarse correlated equilibria. Moreover, convergence results based on generic no-regret algorithms typically use a weaker notion of convergence: the convergence of the average plays instead of the actual plays. Some work has been done showing that when using a specific no-regret algorithm, the well-known multiplicative updates algorithm, convergence of actual plays to equilibria can be shown and better quality of outcomes in terms of the price of anarchy can be reached for atomic congestion games and load balancing games. Are there more cases of natural no-regret dynamics that perform well in suitable classes of games in terms of convergence and quality of outcomes that the dynamics converge to? We answer this question positively in the bulletin-board model by showing that when employing the mirror-descent algorithm, a well-known generic no-regret algorithm, the actual plays converge quickly to equilibria in nonatomic congestion games. Furthermore, the bandit model considers a probably more realistic and prevalent setting with only partial information, in which at each time step each player only knows the cost of her own currently played strategy, but not any costs of unplayed strategies. For the class of atomic congestion games, we propose a family of bandit algorithms based on the mirror-descent algorithms previously presented, and show that when each player individually adopts such a bandit algorithm, their joint (mixed) strategy profile quickly converges with implications.

false

true

false

true

56,341

1811.08919

Robust Active Learning for Electrocardiographic Signal Classification

The classification of electrocardiographic (ECG) signals is a challenging problem for healthcare industry. Traditional supervised learning methods require a large number of labeled data which is usually expensive and difficult to obtain for ECG signals. Active learning is well-suited for ECG signal classification as it aims at selecting the best set of labeled data in order to maximize the classification performance. Motivated by the fact that ECG data are usually heavily unbalanced among different classes and the class labels are noisy as they are manually labeled, this paper proposes a novel solution based on robust active learning for addressing these challenges. The key idea is to first apply the clustering of the data in a low dimensional embedded space and then select the most information instances within local clusters. By selecting the most informative instances relying on local average minimal distances, the algorithm tends to select the data for labelling in a more diversified way. Finally, the robustness of the model is further enhanced by adding a novel noisy label reduction scheme after the selection of the labeled data. Experiments on the ECG signal classification from the MIT-BIH arrhythmia database demonstrate the effectiveness of the proposed algorithm.

false

true

false

114,145

2302.02405

Convergence Analysis of the Deep Galerkin Method for Weak Solutions

This paper analyzes the convergence rate of a deep Galerkin method for the weak solution (DGMW) of second-order elliptic partial differential equations on $\mathbb{R}^d$ with Dirichlet, Neumann, and Robin boundary conditions, respectively. In DGMW, a deep neural network is applied to parametrize the PDE solution, and a second neural network is adopted to parametrize the test function in the traditional Galerkin formulation. By properly choosing the depth and width of these two networks in terms of the number of training samples $n$, it is shown that the convergence rate of DGMW is $\mathcal{O}(n^{-1/d})$, which is the first convergence result for weak solutions. The main idea of the proof is to divide the error of the DGMW into an approximation error and a statistical error. We derive an upper bound on the approximation error in the $H^{1}$ norm and bound the statistical error via Rademacher complexity.

false

true

false

true

343,987

2109.14111

Modeling and Control of bittide Synchronization

Distributed system applications rely on a fine-grain common sense of time. Existing systems maintain the common sense of time by keeping each independent machine as close as possible to wall-clock time through a combination of software protocols like NTP and GPS signals and/or precision references like atomic clocks. This approach is expensive and has tolerance limitations that require protocols to deal with asynchrony and its performance consequences. Moreover, at data-center scale it is impractical to distribute a physical clock as is done on a chip or printed circuit board. In this paper we introduce a distributed system design that removes the need for physical clock distribution or mechanisms for maintaining close alignment to wall-clock time, and instead provides applications with a perfectly synchronized logical clock. We discuss the abstract frame model (AFM), a mathematical model that underpins the system synchronization. The model is based on the rate of communication between nodes in a topology without requiring a global clock. We show that there are families of controllers that satisfy the properties required for existence and uniqueness of solutions to the AFM, and give examples.

false

true

false

true

257,849

2307.00781

ACDMSR: Accelerated Conditional Diffusion Models for Single Image Super-Resolution

Diffusion models have gained significant popularity in the field of image-to-image translation. Previous efforts applying diffusion models to image super-resolution (SR) have demonstrated that iteratively refining pure Gaussian noise using a U-Net architecture trained on denoising at various noise levels can yield satisfactory high-resolution images from low-resolution inputs. However, this iterative refinement process comes with the drawback of low inference speed, which strongly limits its applications. To speed up inference and further enhance the performance, our research revisits diffusion models in image super-resolution and proposes a straightforward yet significant diffusion model-based super-resolution method called ACDMSR (accelerated conditional diffusion model for image super-resolution). Specifically, our method adapts the standard diffusion model to perform super-resolution through a deterministic iterative denoising process. Our study also highlights the effectiveness of using a pre-trained SR model to provide the conditional image of the given low-resolution (LR) image to achieve superior high-resolution results. We demonstrate that our method surpasses previous attempts in qualitative and quantitative results through extensive experiments conducted on benchmark datasets such as Set5, Set14, Urban100, BSD100, and Manga109. Moreover, our approach generates more visually realistic counterparts for low-resolution images, emphasizing its effectiveness in practical scenarios.

false

true

false

377,126

2502.04030

Fine, I'll Merge It Myself: A Multi-Fidelity Framework for Automated Model Merging

Reasoning capabilities represent a critical frontier for large language models (LLMs), but developing them requires extensive proprietary datasets and computational resources. One way to efficiently supplement capabilities with is by model merging, which offers a promising alternative by combining multiple models without retraining. However, current merging approaches rely on manually-designed strategies for merging hyperparameters, limiting the exploration of potential model combinations and requiring significant human effort. We propose an Automated Model Merging Framework that enables fine-grained exploration of merging strategies while reducing costs through multi-fidelity approximations. We support both single and multi-objective optimization and introduce two novel search spaces: layerwise fusion (LFS) and depth-wise integration (DIS). Evaluating across a number of benchmarks, we find that the search autonomously finds 1) Merges that further boost single-objective performance, even on tasks the model has already been finetuned on, and 2) Merges that optimize multi-objective frontiers across tasks. Effective merges are found with limited compute, e.g. within less than 500 search steps.

false

true

false

true

false

530,956

2210.02356

A Liquid Democracy System for Human-Computer Societies

Problem of reliable democratic governance is critical for survival of any community, and it will be critical for communities powered with Artificial Intelligence (AI) systems upon developments of the latter. Apparently, it will be getting more and more critical because of increasing speeds and scales of electronic communications and decreasing latencies in system responses. In order to address this need, we present design and implementation of a reputation system supporting "liquid democracy" principle. The system is based on "weighted liquid rank" algorithm employing different sorts of explicit and implicit ratings being exchanged by members of the society as well as implicit assessments of of the members based on measures of their activity in the society. The system is evaluated against live social network data with help of simulation modelling for an online marketplace case.

false

true

false

true

false

321,620

2210.00094

Improving Robustness with Adaptive Weight Decay

We propose adaptive weight decay, which automatically tunes the hyper-parameter for weight decay during each training iteration. For classification problems, we propose changing the value of the weight decay hyper-parameter on the fly based on the strength of updates from the classification loss (i.e., gradient of cross-entropy), and the regularization loss (i.e., $\ell_2$-norm of the weights). We show that this simple modification can result in large improvements in adversarial robustness -- an area which suffers from robust overfitting -- without requiring extra data across various datasets and architecture choices. For example, our reformulation results in $20\%$ relative robustness improvement for CIFAR-100, and $10\%$ relative robustness improvement on CIFAR-10 comparing to the best tuned hyper-parameters of traditional weight decay resulting in models that have comparable performance to SOTA robustness methods. In addition, this method has other desirable properties, such as less sensitivity to learning rate, and smaller weight norms, which the latter contributes to robustness to overfitting to label noise, and pruning.

false

true

false

true

false

320,729

1803.06492

Evolving Deep Convolutional Neural Networks by Variable-length Particle Swarm Optimization for Image Classification

Convolutional neural networks (CNNs) are one of the most effective deep learning methods to solve image classification problems, but the best architecture of a CNN to solve a specific problem can be extremely complicated and hard to design. This paper focuses on utilising Particle Swarm Optimisation (PSO) to automatically search for the optimal architecture of CNNs without any manual work involved. In order to achieve the goal, three improvements are made based on traditional PSO. First, a novel encoding strategy inspired by computer networks which empowers particle vectors to easily encode CNN layers is proposed; Second, in order to allow the proposed method to learn variable-length CNN architectures, a Disabled layer is designed to hide some dimensions of the particle vector to achieve variable-length particles; Third, since the learning process on large data is slow, partial datasets are randomly picked for the evaluation to dramatically speed it up. The proposed algorithm is examined and compared with 12 existing algorithms including the state-of-art methods on three widely used image classification benchmark datasets. The experimental results show that the proposed algorithm is a strong competitor to the state-of-art algorithms in terms of classification error. This is the first work using PSO for automatically evolving the architectures of CNNs.

false

true

false

true

false

92,845

2407.08260

SALSA: Swift Adaptive Lightweight Self-Attention for Enhanced LiDAR Place Recognition

Large-scale LiDAR mappings and localization leverage place recognition techniques to mitigate odometry drifts, ensuring accurate mapping. These techniques utilize scene representations from LiDAR point clouds to identify previously visited sites within a database. Local descriptors, assigned to each point within a point cloud, are aggregated to form a scene representation for the point cloud. These descriptors are also used to re-rank the retrieved point clouds based on geometric fitness scores. We propose SALSA, a novel, lightweight, and efficient framework for LiDAR place recognition. It consists of a Sphereformer backbone that uses radial window attention to enable information aggregation for sparse distant points, an adaptive self-attention layer to pool local descriptors into tokens, and a multi-layer-perceptron Mixer layer for aggregating the tokens to generate a scene descriptor. The proposed framework outperforms existing methods on various LiDAR place recognition datasets in terms of both retrieval and metric localization while operating in real-time.

false

true

false

true

false

472,094

1908.09067

Plexus Convolutional Neural Network (PlexusNet): A novel neural network architecture for histologic image analysis

Different convolutional neural network (CNN) models have been tested for their application in histological image analyses. However, these models are prone to overfitting due to their large parameter capacity, requiring more data or valuable computational resources for model training. Given these limitations, we introduced a novel architecture (termed PlexusNet). We utilized 310 Hematoxylin and Eosin stained (H&E) annotated histological images of prostate cancer cases from TCGA-PRAD and Stanford University and 398 H&E whole slides images from the Camelyon 2016 challenge. PlexusNet-architecture -derived models were compared to models derived from several existing "state of the art" architectures. We measured discrimination accuracy, calibration, and clinical utility. An ablation study was conducted to study the effect of each component of PlexusNet on model performance. A well-fitted PlexusNet-based model delivered comparable classification performance (AUC: 0.963) in distinguishing prostate cancer from healthy tissues, although it was at least 23 times smaller, had a better model calibration and clinical utility than the comparison models. A separate smaller PlexusNet model accurately detected slides with breast cancer metastases (AUC: 0.978); it helped reduce the slide number to examine by 43.8% without consequences, although its parameter capacity was 200 times smaller than ResNet18. We found that the partitioning of the development set influences the model calibration for all models. However, with PlexusNet architecture, we could achieve comparable well-calibrated models trained on different partitions. In conclusion, PlexusNet represents a novel model architecture for histological image analysis that achieves classification performance comparable to other models while providing orders-of-magnitude parameter reduction.

false

true

false

true

false

142,741

2208.02007

Maintaining Performance with Less Data

We propose a novel method for training a neural network for image classification to reduce input data dynamically, in order to reduce the costs of training a neural network model. As Deep Learning tasks become more popular, their computational complexity increases, leading to more intricate algorithms and models which have longer runtimes and require more input data. The result is a greater cost on time, hardware, and environmental resources. By using data reduction techniques, we reduce the amount of work performed, and therefore the environmental impact of AI techniques, and with dynamic data reduction we show that accuracy may be maintained while reducing runtime by up to 50%, and reducing carbon emission proportionally.

false

true

false

true

false

311,348

1809.01567

Deep Depth from Defocus: how can defocus blur improve 3D estimation using dense neural networks?

Depth estimation is of critical interest for scene understanding and accurate 3D reconstruction. Most recent approaches in depth estimation with deep learning exploit geometrical structures of standard sharp images to predict corresponding depth maps. However, cameras can also produce images with defocus blur depending on the depth of the objects and camera settings. Hence, these features may represent an important hint for learning to predict depth. In this paper, we propose a full system for single-image depth prediction in the wild using depth-from-defocus and neural networks. We carry out thorough experiments to test deep convolutional networks on real and simulated defocused images using a realistic model of blur variation with respect to depth. We also investigate the influence of blur on depth prediction observing model uncertainty with a Bayesian neural network approach. From these studies, we show that out-of-focus blur greatly improves the depth-prediction network performances. Furthermore, we transfer the ability learned on a synthetic, indoor dataset to real, indoor and outdoor images. For this purpose, we present a new dataset containing real all-focus and defocused images from a Digital Single-Lens Reflex (DSLR) camera, paired with ground truth depth maps obtained with an active 3D sensor for indoor scenes. The proposed approach is successfully validated on both this new dataset and standard ones as NYUv2 or Depth-in-the-Wild. Code and new datasets are available at https://github.com/marcelampc/d3net_depth_estimation

false

true

false

106,836

2001.11610

UAV Autonomous Localization using Macro-Features Matching with a CAD Model

Research in the field of autonomous Unmanned Aerial Vehicles (UAVs) has significantly advanced in recent years, mainly due to their relevance in a large variety of commercial, industrial, and military applications. However, UAV navigation in GPS-denied environments continues to be a challenging problem that has been tackled in recent research through sensor-based approaches. This paper presents a novel offline, portable, real-time in-door UAV localization technique that relies on macro-feature detection and matching. The proposed system leverages the support of machine learning, traditional computer vision techniques, and pre-existing knowledge of the environment. The main contribution of this work is the real-time creation of a macro-feature description vector from the UAV captured images which are simultaneously matched with an offline pre-existing vector from a Computer-Aided Design (CAD) model. This results in a quick UAV localization within the CAD model. The effectiveness and accuracy of the proposed system were evaluated through simulations and experimental prototype implementation. Final results reveal the algorithm's low computational burden as well as its ease of deployment in GPS-denied environments.

false

true

false

162,116

2010.14226

Range-Net: A High Precision Streaming SVD for Big Data Applications

In a Big Data setting computing the dominant SVD factors is restrictive due to the main memory requirements. Recently introduced streaming Randomized SVD schemes work under the restrictive assumption that the singular value spectrum of the data has exponential decay. This is seldom true for any practical data. Although these methods are claimed to be applicable to scientific computations due to associated tail-energy error bounds, the approximation errors in the singular vectors and values are high when the aforementioned assumption does not hold. Furthermore from a practical perspective, oversampling can still be memory intensive or worse can exceed the feature dimension of the data. To address these issues, we present Range-Net as an alternative to randomized SVD that satisfies the tail-energy lower bound given by Eckart-Young-Mirsky (EYM) theorem. Range-Net is a deterministic two-stage neural optimization approach with random initialization, where the main memory requirement depends explicitly on the feature dimension and desired rank, independent of the sample dimension. The data samples are read in a streaming setting with the network minimization problem converging to the desired rank-r approximation. Range-Net is fully interpretable where all the network outputs and weights have a specific meaning. We provide theoretical guarantees that Range-Net extracted SVD factors satisfy EYM tail-energy lower bound at machine precision. Our numerical experiments on real data at various scales confirms this bound. A comparison against the state of the art streaming Randomized SVD shows that Range-Net accuracy is better by six orders of magnitude while being memory efficient.

false

true

false

true

false

true

203,378

1504.01004

Managing Multi-Granular Linguistic Distribution Assessments in Large-Scale Multi-Attribute Group Decision Making

Linguistic large-scale group decision making (LGDM) problems are more and more common nowadays. In such problems a large group of decision makers are involved in the decision process and elicit linguistic information that are usually assessed in different linguistic scales with diverse granularity because of decision makers' distinct knowledge and background. To keep maximum information in initial stages of the linguistic LGDM problems, the use of multi-granular linguistic distribution assessments seems a suitable choice, however to manage such multigranular linguistic distribution assessments, it is necessary the development of a new linguistic computational approach. In this paper it is proposed a novel computational model based on the use of extended linguistic hierarchies, which not only can be used to operate with multi-granular linguistic distribution assessments, but also can provide interpretable linguistic results to decision makers. Based on this new linguistic computational model, an approach to linguistic large-scale multi-attribute group decision making is proposed and applied to a talent selection process in universities.

false

true

false

41,751

2201.06518

Structured model order reduction for vibro-acoustic problems using interpolation and balancing methods

Vibration and dissipation in vibro-acoustic systems can be assessed using frequency response analysis. Evaluating a frequency sweep on a full-order model can be very costly, so model order reduction methods are employed to compute cheap-to-evaluate surrogates. This work compares structure-preserving model reduction methods based on rational interpolation and balanced truncation with a specific focus on their applicability to vibro-acoustic systems. Such models typically exhibit a second-order structure and their material properties as well as their excitation may be depending on the driving frequency. We show and compare the effectiveness of all considered methods by applying them to numerical models of vibro-acoustic systems depicting structural vibration, sound transmission, acoustic scattering, and poroelastic problems.

false

true

false

true

275,749

1908.01667

A principled approach for generating adversarial images under non-smooth dissimilarity metrics

Deep neural networks perform well on real world data but are prone to adversarial perturbations: small changes in the input easily lead to misclassification. In this work, we propose an attack methodology not only for cases where the perturbations are measured by $\ell_p$ norms, but in fact any adversarial dissimilarity metric with a closed proximal form. This includes, but is not limited to, $\ell_1, \ell_2$, and $\ell_\infty$ perturbations; the $\ell_0$ counting "norm" (i.e. true sparseness); and the total variation seminorm, which is a (non-$\ell_p$) convolutional dissimilarity measuring local pixel changes. Our approach is a natural extension of a recent adversarial attack method, and eliminates the differentiability requirement of the metric. We demonstrate our algorithm, ProxLogBarrier, on the MNIST, CIFAR10, and ImageNet-1k datasets. We consider undefended and defended models, and show that our algorithm easily transfers to various datasets. We observe that ProxLogBarrier outperforms a host of modern adversarial attacks specialized for the $\ell_0$ case. Moreover, by altering images in the total variation seminorm, we shed light on a new class of perturbations that exploit neighboring pixel information.

false

true

false

true

false

140,816

2312.01042

Covert Communications in STAR-RIS-Aided Rate-Splitting Multiple Access Systems

In this paper, we investigate covert communications in a simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS)-aided rate-splitting multiple access (RSMA) system. Under the RSMA principles, the messages for the covert user (Bob) and public user (Grace) are converted to the common and private streams at the legitimate transmitter (Alice) to realize downlink transmissions, while the STAR-RIS is deployed not only to aid the public transmissions from Alice to Grace, but also to shield the covert transmissions from Alice to Bob against the warden (Willie). To characterize the covert performance of the considered STAR-RIS-aided RSMA (STAR-RIS-RSMA) system, we derive analytical expression for the minimum average detection error probability of Willie, based on which a covert rate maximization problem is formulated. To maximize Bob's covert rate while confusing Willie's monitoring, the transmit power allocation, common rate allocation, and STAR-RIS reflection/transmission beamforming are jointly optimized subject to Grace's quality of service (QoS) requirements. The non-convex covert rate maximization problem, consisting of highly coupled system parameters are decoupled into three sub-problems of transmit power allocation, common rate allocation, and STAR-RIS reflection/transmission beamforming, respectively. To obtain the rank-one constrained optimal solution for the sub-problem of optimizing the STAR-RIS reflection/transmission beamforming, a penalty-based successive convex approximation scheme is developed. Moreover, an alternative optimization (AO) algorithm is designed to determine the optimal solution for the sub-problem of optimizing the transmit power allocation, while the original problem is overall solved by a new AO algorithm.

false

true

false

412,281

2309.04185

Predictive and Robust Robot Assistance for Sequential Manipulation

This paper presents a novel concept to support physically impaired humans in daily object manipulation tasks with a robot. Given a user's manipulation sequence, we propose a predictive model that uniquely casts the user's sequential behavior as well as a robot support intervention into a hierarchical multi-objective optimization problem. A major contribution is the prediction formulation, which allows to consider several different future paths concurrently. The second contribution is the encoding of a general notion of constancy constraints, which allows to consider dependencies between consecutive or far apart keyframes (in time or space) of a sequential task. We perform numerical studies, simulations and robot experiments to analyse and evaluate the proposed method in several table top tasks where a robot supports impaired users by predicting their posture and proactively re-arranging objects.

false

true

false

390,641

1807.08291

Correlation Net: Spatiotemporal multimodal deep learning for action recognition

This paper describes a network that captures multimodal correlations over arbitrary timestamps. The proposed scheme operates as a complementary, extended network over a multimodal convolutional neural network (CNN). Spatial and temporal streams are required for action recognition by a deep CNN, but overfitting reduction and fusing these two streams remain open problems. The existing fusion approach averages the two streams. Here we propose a correlation network with a Shannon fusion for learning a pre-trained CNN. A Long-range video may consist of spatiotemporal correlations over arbitrary times, which can be captured by forming the correlation network from simple fully connected layers. This approach was found to complement the existing network fusion methods. The importance of multimodal correlation is validated in comparison experiments on the UCF-101 and HMDB-51 datasets. The multimodal correlation enhanced the accuracy of the video recognition results.

false

true

false

103,507

1805.02285

Clustering With Pairwise Relationships: A Generative Approach

Semi-supervised learning (SSL) has become important in current data analysis applications, where the amount of unlabeled data is growing exponentially and user input remains limited by logistics and expense. Constrained clustering, as a subclass of SSL, makes use of user input in the form of relationships between data points (e.g., pairs of data points belonging to the same class or different classes) and can remarkably improve the performance of unsupervised clustering in order to reflect user-defined knowledge of the relationships between particular data points. Existing algorithms incorporate such user input, heuristically, as either hard constraints or soft penalties, which are separate from any generative or statistical aspect of the clustering model; this results in formulations that are suboptimal and not sufficiently general. In this paper, we propose a principled, generative approach to probabilistically model, without ad hoc penalties, the joint distribution given by user-defined pairwise relations. The proposed model accounts for general underlying distributions without assuming a specific form and relies on expectation-maximization for model fitting. For distributions in a standard form, the proposed approach results in a closed-form solution for updated parameters.

false

true

false

96,828

1506.05158

An Entropy Maximizing Geohash for Distributed Spatiotemporal Database Indexing

We present a modification of the standard geohash algorithm based on maximum entropy encoding in which the data volume is approximately constant for a given hash prefix length. Distributed spatiotemporal databases, which typically require interleaving spatial and temporal elements into a single key, reap large benefits from a balanced geohash by creating a consistent ratio between spatial and temporal precision even across areas of varying data density. This property is also useful for indexing purely spatial datasets, where the load distribution of large range scans is an important aspect of query performance. We apply our algorithm to data generated proportional to population as given by census block population counts provided from the US Census Bureau.

false

true

44,260

1503.03131

Discovering functional zones using bus smart card data and points of interest in Beijing

Cities comprise various functional zones, including residential, educational, commercial zones, etc. It is important for urban planners to identify different functional zones and understand their spatial structure within the city in order to make better urban plans. In this research, we used 77976010 bus smart card records of Beijing City in one week in April 2008 and converted them into two-dimensional time series data of each bus platform, Then, through data mining in the big database system and previous studies on citizens' trip behavior, we established the DZoF (discovering zones of different functions) model based on SCD (smart card Data) and POIs (points of interest), and pooled the results at the TAZ (traffic analysis zone) level. The results suggested that DzoF model and cluster analysis based on dimensionality reduction and EM (expectation-maximization) algorithm can identify functional zones that well match the actual land uses in Beijing. The methodology in the present research can help urban planners and the public understand the complex urban spatial structure and contribute to the academia of urban geography and urban planning.

false

true

false

true

false

41,012

2111.08452

On minimizers and convolutional filters: theoretical connections and applications to genome analysis

Minimizers and convolutional neural networks (CNNs) are two quite distinct popular techniques that have both been employed to analyze categorical biological sequences. At face value, the methods seem entirely dissimilar. Minimizers use min-wise hashing on a rolling window to extract a single important k-mer feature per window. CNNs start with a wide array of randomly initialized convolutional filters, paired with a pooling operation, and then multiple additional neural layers to learn both the filters themselves and how they can be used to classify the sequence. Here, our main result is a careful mathematical analysis of hash function properties showing that for sequences over a categorical alphabet, random Gaussian initialization of convolutional filters with max-pooling is equivalent to choosing a minimizer ordering such that selected k-mers are (in Hamming distance) far from the k-mers within the sequence but close to other minimizers. In empirical experiments, we find that this property manifests as decreased density in repetitive regions, both in simulation and on real human telomeres. We additionally train from scratch a CNN embedding of synthetic short-reads from the SARS-CoV-2 genome into 3D Euclidean space that locally recapitulates the linear sequence distance of the read origins, a modest step towards building a deep learning assembler, though it is at present too slow to be practical. In total, this manuscript provides a partial explanation for the effectiveness of CNNs in categorical sequence analysis.

false

true

false

true

false

266,695

2106.02605

A Holistic Approach to Interpretability in Financial Lending: Models, Visualizations, and Summary-Explanations

Lending decisions are usually made with proprietary models that provide minimally acceptable explanations to users. In a future world without such secrecy, what decision support tools would one want to use for justified lending decisions? This question is timely, since the economy has dramatically shifted due to a pandemic, and a massive number of new loans will be necessary in the short term. We propose a framework for such decisions, including a globally interpretable machine learning model, an interactive visualization of it, and several types of summaries and explanations for any given decision. The machine learning model is a two-layer additive risk model, which resembles a two-layer neural network, but is decomposable into subscales. In this model, each node in the first (hidden) layer represents a meaningful subscale model, and all of the nonlinearities are transparent. Our online visualization tool allows exploration of this model, showing precisely how it came to its conclusion. We provide three types of explanations that are simpler than, but consistent with, the global model: case-based reasoning explanations that use neighboring past cases, a set of features that were the most important for the model's prediction, and summary-explanations that provide a customized sparse explanation for any particular lending decision made by the model. Our framework earned the FICO recognition award for the Explainable Machine Learning Challenge, which was the first public challenge in the domain of explainable machine learning.

false

true

false

238,941

2304.00212

Devil is in the Queries: Advancing Mask Transformers for Real-world Medical Image Segmentation and Out-of-Distribution Localization

Real-world medical image segmentation has tremendous long-tailed complexity of objects, among which tail conditions correlate with relatively rare diseases and are clinically significant. A trustworthy medical AI algorithm should demonstrate its effectiveness on tail conditions to avoid clinically dangerous damage in these out-of-distribution (OOD) cases. In this paper, we adopt the concept of object queries in Mask Transformers to formulate semantic segmentation as a soft cluster assignment. The queries fit the feature-level cluster centers of inliers during training. Therefore, when performing inference on a medical image in real-world scenarios, the similarity between pixels and the queries detects and localizes OOD regions. We term this OOD localization as MaxQuery. Furthermore, the foregrounds of real-world medical images, whether OOD objects or inliers, are lesions. The difference between them is less than that between the foreground and background, possibly misleading the object queries to focus redundantly on the background. Thus, we propose a query-distribution (QD) loss to enforce clear boundaries between segmentation targets and other regions at the query level, improving the inlier segmentation and OOD indication. Our proposed framework is tested on two real-world segmentation tasks, i.e., segmentation of pancreatic and liver tumors, outperforming previous state-of-the-art algorithms by an average of 7.39% on AUROC, 14.69% on AUPR, and 13.79% on FPR95 for OOD localization. On the other hand, our framework improves the performance of inlier segmentation by an average of 5.27% DSC when compared with the leading baseline nnUNet.

false

true

false

true

false

355,598

1811.08290

An Efficient Optical Flow Based Motion Detection Method for Non-stationary Scenes

Real-time motion detection in non-stationary scenes is a difficult task due to dynamic background, changing foreground appearance and limited computational resource. These challenges degrade the performance of the existing methods in practical applications. In this paper, an optical flow based framework is proposed to address this problem. By applying a novel strategy to utilize optical flow, we enable our method being free of model constructing, training or updating and can be performed efficiently. Besides, a dual judgment mechanism with adaptive intervals and adaptive thresholds is designed to heighten the system's adaptation to different situations. In experiment part, we quantitatively and qualitatively validate the effectiveness and feasibility of our method with videos in various scene conditions. The experimental results show that our method adapts itself to different situations and outperforms the state-of-the-art real-time methods, indicating the advantages of our optical flow based method.

false

true

false

114,005

2402.16328

A Joint Communication and Computation Design for Probabilistic Semantic Communications

In this paper, the problem of joint transmission and computation resource allocation for a multi-user probabilistic semantic communication (PSC) network is investigated. In the considered model, users employ semantic information extraction techniques to compress their large-sized data before transmitting them to a multi-antenna base station (BS). Our model represents large-sized data through substantial knowledge graphs, utilizing shared probability graphs between the users and the BS for efficient semantic compression. The resource allocation problem is formulated as an optimization problem with the objective of maximizing the sum of equivalent rate of all users, considering total power budget and semantic resource limit constraints. The computation load considered in the PSC network is formulated as a non-smooth piecewise function with respect to the semantic compression ratio. To tackle this non-convex non-smooth optimization challenge, a three-stage algorithm is proposed where the solutions for the receive beamforming matrix of the BS, transmit power of each user, and semantic compression ratio of each user are obtained stage by stage. Numerical results validate the effectiveness of our proposed scheme.

false

true

false

432,519

2408.06235

Correlation Weighted Prototype-based Self-Supervised One-Shot Segmentation of Medical Images

Medical image segmentation is one of the domains where sufficient annotated data is not available. This necessitates the application of low-data frameworks like few-shot learning. Contemporary prototype-based frameworks often do not account for the variation in features within the support and query images, giving rise to a large variance in prototype alignment. In this work, we adopt a prototype-based self-supervised one-way one-shot learning framework using pseudo-labels generated from superpixels to learn the semantic segmentation task itself. We use a correlation-based probability score to generate a dynamic prototype for each query pixel from the bag of prototypes obtained from the support feature map. This weighting scheme helps to give a higher weightage to contextually related prototypes. We also propose a quadrant masking strategy in the downstream segmentation task by utilizing prior domain information to discard unwanted false positives. We present extensive experimentations and evaluations on abdominal CT and MR datasets to show that the proposed simple but potent framework performs at par with the state-of-the-art methods.

false

true

false

480,122

2407.06915

FE-GUT: Factor Graph Optimization hybrid with Extended Kalman Filter for tightly coupled GNSS/UWB Integration

Precise positioning and navigation information has been increasingly important with the development of the consumer electronics market. Due to some deficits of Global Navigation Satellite System (GNSS), such as susceptible to interferences, integrating of GNSS with additional alternative sensors is a promising approach to overcome the performance limitations of GNSS-based localization systems. Ultra-Wideband (UWB) can be used to enhance GNSS in constructing an integrated localization system. However, most low-cost UWB devices lack a hardware-level time synchronization feature, which necessitates the estimation and compensation of the time-offset in the tightly coupled GNSS/UWB integration. Given the flexibility of probabilistic graphical models, the time-offset can be modeled as an invariant constant in the discretization of the continuous model. This work proposes a novel architecture in which Factor Graph Optimization (FGO) is hybrid with Extend Kalman Filter (EKF) for tightly coupled GNSS/UWB integration with online Temporal calibration (FE-GUT). FGO is utilized to precisely estimate the time-offset, while EKF provides initailization for the new factors and performs time-offset compensation. Simulation-based experiments validate the integrated localization performance of FE-GUT. In a four-wheeled robot scenario, the results demonstrate that, compared to EKF, FE-GUT can improve horizontal and vertical localization accuracy by 58.59\% and 34.80\%, respectively, while the time-offset estimation accuracy is improved by 76.80\%. All the source codes and datasets can be gotten via https://github.com/zhaoqj23/FE-GUT/.

false

true

false

471,576

1805.09233

Segmentation of Liver Lesions with Reduced Complexity Deep Models

We propose a computationally efficient architecture that learns to segment lesions from CT images of the liver. The proposed architecture uses bilinear interpolation with sub-pixel convolution at the last layer to upscale the course feature in bottle neck architecture. Since bilinear interpolation and sub-pixel convolution do not have any learnable parameter, our overall model is faster and occupies less memory footprint than the traditional U-net. We evaluate our proposed architecture on the highly competitive dataset of 2017 Liver Tumor Segmentation (LiTS) Challenge. Our method achieves competitive results while reducing the number of learnable parameters roughly by a factor of 13.8 compared to the original UNet model.

false

true

false

98,374

1601.00042

Fast, Safe, and Propellant-Efficient Spacecraft Planning under Clohessy-Wiltshire-Hill Dynamics

This paper presents a sampling-based motion planning algorithm for real-time and propellant-optimized autonomous spacecraft trajectory generation in near-circular orbits. Specifically, this paper leverages recent algorithmic advances in the field of robot motion planning to the problem of impulsively-actuated, propellant-optimized rendezvous and proximity operations under the Clohessy-Wiltshire-Hill (CWH) dynamics model. The approach calls upon a modified version of the Fast Marching Tree (FMT*) algorithm to grow a set of feasible trajectories over a deterministic, low-dispersion set of sample points covering the free state space. To enforce safety, the tree is only grown over the subset of actively-safe samples, from which there exists a feasible one-burn collision avoidance maneuver that can safely circularize the spacecraft orbit along its coasting arc under a given set of potential thruster failures. Key features of the proposed algorithm include: (i) theoretical guarantees in terms of trajectory safety and performance, (ii) amenability to real-time implementation, and (iii) generality, in the sense that a large class of constraints can be handled directly. As a result, the proposed algorithm offers the potential for widespread application, ranging from on-orbit satellite servicing to orbital debris removal and autonomous inspection missions.

false

true

false

50,598

2407.11698

NITRO-D: Native Integer-only Training of Deep Convolutional Neural Networks

Quantization has become increasingly pivotal in addressing the steadily increasing computational and memory requirements of Deep Neural Networks (DNNs). By reducing the number of bits used to represent weights and activations (typically from 32-bit floating-point to 16-bit or 8-bit integers), quantization reduces the memory footprint, energy consumption, and execution time of DNN models. However, traditional quantization methods typically focus on the inference of DNNs, while the training process still relies on floating-point operations. To date, only one work in the literature has addressed integer-only training for Multi-Layer Perceptron (MLP) architectures. This work introduces NITRO-D, a new framework for training arbitrarily deep integer-only Convolutional Neural Networks (CNNs) that operate entirely in the integer-only domain for both training and inference. NITRO-D is the first framework in the literature enabling the training of integer-only CNNs without the need to introduce a quantization scheme. Specifically, NITRO-D introduces a novel architecture integrating multiple integer local-loss blocks, which include the proposed NITRO Scaling Layer and the NITRO-ReLU activation function. Additionally, it introduces a novel integer-only learning algorithm derived from Local Error Signals (LES), utilizing IntegerSGD, an optimizer specifically designed to operate in an integer-only context. NITRO-D is implemented in an open-source Python library. Extensive experimental evaluations demonstrate its effectiveness across several state-of-the-art image recognition datasets. Results show significant performance improvements from 2.47% to 5.96% for integer-only MLP architectures over the state-of-the-art solution, and the capability of training integer-only CNN architectures with minimal accuracy degradation from -0.15% to -4.22% compared to floating-point LES.

false

true

false

true

false

true

false

true

false

473,580

2408.02544

Caution for the Environment: Multimodal Agents are Susceptible to Environmental Distractions

This paper investigates the faithfulness of multimodal large language model (MLLM) agents in the graphical user interface (GUI) environment, aiming to address the research question of whether multimodal GUI agents can be distracted by environmental context. A general setting is proposed where both the user and the agent are benign, and the environment, while not malicious, contains unrelated content. A wide range of MLLMs are evaluated as GUI agents using our simulated dataset, following three working patterns with different levels of perception. Experimental results reveal that even the most powerful models, whether generalist agents or specialist GUI agents, are susceptible to distractions. While recent studies predominantly focus on the helpfulness (i.e., action accuracy) of multimodal agents, our findings indicate that these agents are prone to environmental distractions, resulting in unfaithful behaviors. Furthermore, we switch to the adversarial perspective and implement environment injection, demonstrating that such unfaithfulness can be exploited, leading to unexpected risks.

false

true

false

478,670

1702.05779

Evaluation of Lane Departure Correction Systems Using a Stochastic Driver Model

Evaluating the effectiveness and benefits of driver assistance systems is crucial for improving the system performance. In this paper, we propose a novel framework for testing and evaluating lane departure correction systems at a low cost by using lane departure events reproduced from naturalistic driving data. First, 529,096 lane departure events were extracted from the Safety Pilot Model Deployment (SPMD) database collected by the University of Michigan Transportation Research Institute. Second, a stochastic lane departure model consisting of eight random key variables was developed to reduce the dimension of the data description and improve the computational efficiency. As such, we used a bounded Gaussian mixture model (BGM) model to describe drivers' stochastic lane departure behaviors. Then, a lane departure correction system with an aim point controller was designed, and a batch of lane departure events were reproduced from the learned stochastic driver model. Finally, we assessed the developed evaluation approach by comparing lateral departure areas of vehicles between with and without correction controller. The simulation results show that the proposed method can effectively evaluate lane departure correction systems.

false

true

false

68,469

2211.00676

Towards Inter-character Relationship-driven Story Generation

In this paper, we introduce the task of modeling interpersonal relationships for story generation. For addressing this task, we propose Relationships as Latent Variables for Story Generation, (ReLiSt). ReLiSt generates stories sentence by sentence and has two major components - a relationship selector and a story continuer. The relationship selector specifies a latent variable to pick the relationship to exhibit in the next sentence and the story continuer generates the next sentence while expressing the selected relationship in a coherent way. Our automatic and human evaluations demonstrate that ReLiSt is able to generate stories with relationships that are more faithful to desired relationships while maintaining the content quality. The relationship assignments to sentences during inference bring interpretability to ReLiSt.

false

true

false

327,952

1609.02593

Multiple predator based capture process on complex networks

The predator/prey (capture) problem is a prototype of many network-related applications. We study the capture process on complex networks by considering multiple predators from multiple sources. In our model, some lions start from multiple sources simultaneously to capture the lamb by biased random walks, which are controlled with a free parameter $\alpha$. We derive the distribution of the lamb's lifetime and the expected lifetime $\left\langle T\right\rangle $. Through simulation, we find that the expected lifetime drops substantially with the increasing number of lions. We also study how the underlying topological structure affects the capture process, and obtain that locating on small-degree nodes is better than large-degree nodes to prolong the lifetime of the lamb. Moreover, dense or homogeneous network structures are against the survival of the lamb.

false

true

false

60,754

2204.00260

MS-HLMO: Multi-scale Histogram of Local Main Orientation for Remote Sensing Image Registration

Multi-source image registration is challenging due to intensity, rotation, and scale differences among the images. Considering the characteristics and differences of multi-source remote sensing images, a feature-based registration algorithm named Multi-scale Histogram of Local Main Orientation (MS-HLMO) is proposed. Harris corner detection is first adopted to generate feature points. The HLMO feature of each Harris feature point is extracted on a Partial Main Orientation Map (PMOM) with a Generalized Gradient Location and Orientation Histogram-like (GGLOH) feature descriptor, which provides high intensity, rotation, and scale invariance. The feature points are matched through a multi-scale matching strategy. Comprehensive experiments on 17 multi-source remote sensing scenes demonstrate that the proposed MS-HLMO and its simplified version MS-HLMO$^+$ outperform other competitive registration algorithms in terms of effectiveness and generalization.

false

true

false

289,193

2212.09420

Large Language Models Meet NL2Code: A Survey

The task of generating code from a natural language description, or NL2Code, is considered a pressing and significant challenge in code intelligence. Thanks to the rapid development of pre-training techniques, surging large language models are being proposed for code, sparking the advances in NL2Code. To facilitate further research and applications in this field, in this paper, we present a comprehensive survey of 27 existing large language models for NL2Code, and also review benchmarks and metrics. We provide an intuitive comparison of all existing models on the HumanEval benchmark. Through in-depth observation and analysis, we provide some insights and conclude that the key factors contributing to the success of large language models for NL2Code are "Large Size, Premium Data, Expert Tuning". In addition, we discuss challenges and opportunities regarding the gap between models and humans. We also create a website https://nl2code.github.io to track the latest progress through crowd-sourcing. To the best of our knowledge, this is the first survey of large language models for NL2Code, and we believe it will contribute to the ongoing development of the field.

false

true

false

true

false

true

337,105

1811.01294

Learning Contextual Hierarchical Structure of Medical Concepts with Poincair\'e Embeddings to Clarify Phenotypes

Biomedical association studies are increasingly done using clinical concepts, and in particular diagnostic codes from clinical data repositories as phenotypes. Clinical concepts can be represented in a meaningful, vector space using word embedding models. These embeddings allow for comparison between clinical concepts or for straightforward input to machine learning models. Using traditional approaches, good representations require high dimensionality, making downstream tasks such as visualization more difficult. We applied Poincar\'e embeddings in a 2-dimensional hyperbolic space to a large-scale administrative claims database and show performance comparable to 100-dimensional embeddings in a euclidean space. We then examine disease relationships under different disease contexts to better understand potential phenotypes.

false

true

false

112,323

2406.01629

RecDiff: Diffusion Model for Social Recommendation

Social recommendation has emerged as a powerful approach to enhance personalized recommendations by leveraging the social connections among users, such as following and friend relations observed in online social platforms. The fundamental assumption of social recommendation is that socially-connected users exhibit homophily in their preference patterns. This means that users connected by social ties tend to have similar tastes in user-item activities, such as rating and purchasing. However, this assumption is not always valid due to the presence of irrelevant and false social ties, which can contaminate user embeddings and adversely affect recommendation accuracy. To address this challenge, we propose a novel diffusion-based social denoising framework for recommendation (RecDiff). Our approach utilizes a simple yet effective hidden-space diffusion paradigm to alleivate the noisy effect in the compressed and dense representation space. By performing multi-step noise diffusion and removal, RecDiff possesses a robust ability to identify and eliminate noise from the encoded user representations, even when the noise levels vary. The diffusion module is optimized in a downstream task-aware manner, thereby maximizing its ability to enhance the recommendation process. We conducted extensive experiments to evaluate the efficacy of our framework, and the results demonstrate its superiority in terms of recommendation accuracy, training efficiency, and denoising effectiveness. The source code for the model implementation is publicly available at: https://github.com/HKUDS/RecDiff.

false

true

false

460,406

2401.05385

Angle-Equivariant Convolutional Neural Networks for Interference Mitigation in Automotive Radar

In automotive applications, frequency modulated continuous wave (FMCW) radar is an established technology to determine the distance, velocity and angle of objects in the vicinity of the vehicle. The quality of predictions might be seriously impaired if mutual interference between radar sensors occurs. Previous work processes data from the entire receiver array in parallel to increase interference mitigation quality using neural networks (NNs). However, these architectures do not generalize well across different angles of arrival (AoAs) of interferences and objects. In this paper we introduce fully convolutional neural network (CNN) with rank-three convolutions which is able to transfer learned patterns between different AoAs. Our proposed architecture outperforms previous work while having higher robustness and a lower number of trainable parameters. We evaluate our network on a diverse data set and demonstrate its angle equivariance.

false

true

false

420,748

1906.01926

A Resource-Free Evaluation Metric for Cross-Lingual Word Embeddings Based on Graph Modularity

Cross-lingual word embeddings encode the meaning of words from different languages into a shared low-dimensional space. An important requirement for many downstream tasks is that word similarity should be independent of language - i.e., word vectors within one language should not be more similar to each other than to words in another language. We measure this characteristic using modularity, a network measurement that measures the strength of clusters in a graph. Modularity has a moderate to strong correlation with three downstream tasks, even though modularity is based only on the structure of embeddings and does not require any external resources. We show through experiments that modularity can serve as an intrinsic validation metric to improve unsupervised cross-lingual word embeddings, particularly on distant language pairs in low-resource settings.

false

true

false

133,885

2403.13756

Enhancing Gait Video Analysis in Neurodegenerative Diseases by Knowledge Augmentation in Vision Language Model

We present a knowledge augmentation strategy for assessing the diagnostic groups and gait impairment from monocular gait videos. Based on a large-scale pre-trained Vision Language Model (VLM), our model learns and improves visual, textual, and numerical representations of patient gait videos, through a collective learning across three distinct modalities: gait videos, class-specific descriptions, and numerical gait parameters. Our specific contributions are two-fold: First, we adopt a knowledge-aware prompt tuning strategy to utilize the class-specific medical description in guiding the text prompt learning. Second, we integrate the paired gait parameters in the form of numerical texts to enhance the numeracy of the textual representation. Results demonstrate that our model not only significantly outperforms state-of-the-art methods in video-based classification tasks but also adeptly decodes the learned class-specific text features into natural language descriptions using the vocabulary of quantitative gait parameters. The code and the model will be made available at our project page: https://lisqzqng.github.io/GaitAnalysisVLM/.

false

true

false

439,762

2412.15698

Concept Boundary Vectors

Machine learning models are trained with relatively simple objectives, such as next token prediction. However, on deployment, they appear to capture a more fundamental representation of their input data. It is of interest to understand the nature of these representations to help interpret the model's outputs and to identify ways to improve the salience of these representations. Concept vectors are constructions aimed at attributing concepts in the input data to directions, represented by vectors, in the model's latent space. In this work, we introduce concept boundary vectors as a concept vector construction derived from the boundary between the latent representations of concepts. Empirically we demonstrate that concept boundary vectors capture a concept's semantic meaning, and we compare their effectiveness against concept activation vectors.

false

true

false

519,240

1908.05727

Persistent Surveillance With Energy-Constrained UAVs and Mobile Charging Stations

We address the problem of achieving persistent surveillance over an environment by using energy-constrained unmanned aerial vehicles (UAVs), which are supported by unmanned ground vehicles (UGVs) serving as mobile charging stations. Specifically, we plan the trajectories of all vehicles and the charging schedule of UAVs to minimize the long-term maximum age, where age is defined as the time between two consecutive visits to regions of interest in a partitioned environment. We introduce a scalable planning strategy based on 1) creating UAV- UGV teams, 2) decomposing the environment into optimal partitions that can be covered by any of the teams in a single fuel cycle, 3) uniformly distributing the teams over a cyclic path traversing those partitions, and 4) having the UAVs in each team cover their current partition and be transported to the next partition while being recharged by the UGV. We show some results related to the safety and performance of the proposed strategy.

false

true

false

true

false

141,791

1910.11525

Unsupervised Space-Time Clustering using Persistent Homology

This paper presents a new clustering algorithm for space-time data based on the concepts of topological data analysis and in particular, persistent homology. Employing persistent homology - a flexible mathematical tool from algebraic topology used to extract topological information from data - in unsupervised learning is an uncommon and a novel approach. A notable aspect of this methodology consists in analyzing data at multiple resolutions which allows to distinguish true features from noise based on the extent of their persistence. We evaluate the performance of our algorithm on synthetic data and compare it to other well-known clustering algorithms such as K-means, hierarchical clustering and DBSCAN. We illustrate its application in the context of a case study of water quality in the Chesapeake Bay.

false

true

false

150,809

1910.11272

Learning-based real-time method to looking through scattering medium beyond the memory effect

Strong scattering medium brings great difficulties to optical imaging, which is also a problem in medical imaging and many other fields. Optical memory effect makes it possible to image through strong random scattering medium. However, this method also has the limitation of limited angle field-of-view (FOV), which prevents it from being applied in practice. In this paper, a kind of practical convolutional neural network called PDSNet is proposed, which effectively breaks through the limitation of optical memory effect on FOV. Experiments is conducted to prove that the scattered pattern can be reconstructed accurately in real-time by PDSNet, and it is widely applicable to retrieve complex objects of random scales and different scattering media.

false

true

false

150,732

cs/0011018

Optimal Buy-and-Hold Strategies for Financial Markets with Bounded Daily Returns

In the context of investment analysis, we formulate an abstract online computing problem called a planning game and develop general tools for solving such a game. We then use the tools to investigate a practical buy-and-hold trading problem faced by long-term investors in stocks. We obtain the unique optimal static online algorithm for the problem and determine its exact competitive ratio. We also compare this algorithm with the popular dollar averaging strategy using actual market data.

false

true

false

true

537,253

2410.15737

Who's Who: Large Language Models Meet Knowledge Conflicts in Practice

Retrieval-augmented generation (RAG) methods are viable solutions for addressing the static memory limits of pre-trained language models. Nevertheless, encountering conflicting sources of information within the retrieval context is an inevitable practical challenge. In such situations, the language models are recommended to transparently inform users about the conflicts rather than autonomously deciding what to present based on their inherent biases. To analyze how current large language models (LLMs) align with our recommendation, we introduce WhoQA, a public benchmark dataset to examine model's behavior in knowledge conflict situations. We induce conflicts by asking about a common property among entities having the same name, resulting in questions with up to 8 distinctive answers. WhoQA evaluation set includes 5K questions across 13 Wikidata property types and 150K Wikipedia entities. Our experiments show that despite the simplicity of WhoQA questions, knowledge conflicts significantly degrades LLMs' performance in RAG settings.

false

true

false

true

false

500,705

2412.11851

A Benchmark and Robustness Study of In-Context-Learning with Large Language Models in Music Entity Detection

Detecting music entities such as song titles or artist names is a useful application to help use cases like processing music search queries or analyzing music consumption on the web. Recent approaches incorporate smaller language models (SLMs) like BERT and achieve high results. However, further research indicates a high influence of entity exposure during pre-training on the performance of the models. With the advent of large language models (LLMs), these outperform SLMs in a variety of downstream tasks. However, researchers are still divided if this is applicable to tasks like entity detection in texts due to issues like hallucination. In this paper, we provide a novel dataset of user-generated metadata and conduct a benchmark and a robustness study using recent LLMs with in-context-learning (ICL). Our results indicate that LLMs in the ICL setting yield higher performance than SLMs. We further uncover the large impact of entity exposure on the best performing LLM in our study.

false

true

false

true

517,615

cmp-lg/9504005

Constraint Logic Programming for Natural Language Processing

This paper proposes an evaluation of the adequacy of the constraint logic programming paradigm for natural language processing. Theoretical aspects of this question have been discussed in several works. We adopt here a pragmatic point of view and our argumentation relies on concrete solutions. Using actual contraints (in the CLP sense) is neither easy nor direct. However, CLP can improve parsing techniques in several aspects such as concision, control, efficiency or direct representation of linguistic formalism. This discussion is illustrated by several examples and the presentation of an HPSG parser.

false

true

false

536,330

2304.02721

To Asymmetry and Beyond: Structured Pruning of Sequence to Sequence Models for Improved Inference Efficiency

Sequence-to-sequence language models can be used to produce abstractive summaries which are coherent, relevant, and concise. Still, model sizes can make deployment in latency-sensitive or web-scale implementations difficult. This paper studies the relationship between model size, structured pruning, inference efficiency, and summarization accuracy on widely used summarization datasets. We show that model accuracy is tied to the encoder size while inference efficiency is connected to the decoder. Using asymmetric pruning can lead to nearly 3x improvement in inference latency with ~1 point loss in Rouge-2. Moreover, we find both the average degradation and the role of asymmetry to be consistent across model sizes and variations in datasets.

false

true

false

true

false

356,519

2206.07682

Emergent Abilities of Large Language Models

Scaling up language models has been shown to predictably improve performance and sample efficiency on a wide range of downstream tasks. This paper instead discusses an unpredictable phenomenon that we refer to as emergent abilities of large language models. We consider an ability to be emergent if it is not present in smaller models but is present in larger models. Thus, emergent abilities cannot be predicted simply by extrapolating the performance of smaller models. The existence of such emergence implies that additional scaling could further expand the range of capabilities of language models.

false

true

false

302,840

2006.07203

Video Understanding as Machine Translation

With the advent of large-scale multimodal video datasets, especially sequences with audio or transcribed speech, there has been a growing interest in self-supervised learning of video representations. Most prior work formulates the objective as a contrastive metric learning problem between the modalities. To enable effective learning, however, these strategies require a careful selection of positive and negative samples often combined with hand-designed curriculum policies. In this work we remove the need for negative sampling by taking a generative modeling approach that poses the objective as a translation problem between modalities. Such a formulation allows us to tackle a wide variety of downstream video understanding tasks by means of a single unified framework, without the need for large batches of negative samples common in contrastive metric learning. We experiment with the large-scale HowTo100M dataset for training, and report performance gains over the state-of-the-art on several downstream tasks including video classification (EPIC-Kitchens), question answering (TVQA), captioning (TVC, YouCook2, and MSR-VTT), and text-based clip retrieval (YouCook2 and MSR-VTT).

false

true

false

181,717

2408.04673

AutoFAIR : Automatic Data FAIRification via Machine Reading

The explosive growth of data fuels data-driven research, facilitating progress across diverse domains. The FAIR principles emerge as a guiding standard, aiming to enhance the findability, accessibility, interoperability, and reusability of data. However, current efforts primarily focus on manual data FAIRification, which can only handle targeted data and lack efficiency. To address this issue, we propose AutoFAIR, an architecture designed to enhance data FAIRness automately. Firstly, We align each data and metadata operation with specific FAIR indicators to guide machine-executable actions. Then, We utilize Web Reader to automatically extract metadata based on language models, even in the absence of structured data webpage schemas. Subsequently, FAIR Alignment is employed to make metadata comply with FAIR principles by ontology guidance and semantic matching. Finally, by applying AutoFAIR to various data, especially in the field of mountain hazards, we observe significant improvements in findability, accessibility, interoperability, and reusability of data. The FAIRness scores before and after applying AutoFAIR indicate enhanced data value.

false

true

false

true

false

true

false

479,483

1701.05084

Analysis of the noise in back-projection light field acquisition and its optimization

Light field reconstruction from images captured by focal plane sweeping can achieve high lateral resolution comparable to the modern camera sensor. This is impossible for the conventional micro-lenslet based light field capture systems. However, the severe defocus noise and the low depth resolution limit its applications. In this paper, we analyze the defocus noise and the depth resolution in the focal plane sweeping based light field reconstruction technique, and propose a method to reduce the defocus noise and improve the depth resolution. Both numerical and experimental results verify the proposed method.

false

true

false

66,938

2308.15966

SHARP Challenge 2023: Solving CAD History and pArameters Recovery from Point clouds and 3D scans. Overview, Datasets, Metrics, and Baselines

Recent breakthroughs in geometric Deep Learning (DL) and the availability of large Computer-Aided Design (CAD) datasets have advanced the research on learning CAD modeling processes and relating them to real objects. In this context, 3D reverse engineering of CAD models from 3D scans is considered to be one of the most sought-after goals for the CAD industry. However, recent efforts assume multiple simplifications limiting the applications in real-world settings. The SHARP Challenge 2023 aims at pushing the research a step closer to the real-world scenario of CAD reverse engineering through dedicated datasets and tracks. In this paper, we define the proposed SHARP 2023 tracks, describe the provided datasets, and propose a set of baseline methods along with suitable evaluation metrics to assess the performance of the track solutions. All proposed datasets along with useful routines and the evaluation metrics are publicly available.

false

true

false

388,864

cs/0104017

Local Search Techniques for Constrained Portfolio Selection Problems

We consider the problem of selecting a portfolio of assets that provides the investor a suitable balance of expected return and risk. With respect to the seminal mean-variance model of Markowitz, we consider additional constraints on the cardinality of the portfolio and on the quantity of individual shares. Such constraints better capture the real-world trading system, but make the problem more difficult to be solved with exact methods. We explore the use of local search techniques, mainly tabu search, for the portfolio selection problem. We compare and combine previous work on portfolio selection that makes use of the local search approach and we propose new algorithms that combine different neighborhood relations. In addition, we show how the use of randomization and of a simple form of adaptiveness simplifies the setting of a large number of critical parameters. Finally, we show how our techniques perform on public benchmarks.

false

true

false

true

false

537,318

2301.02754

On Frequency-Based Optimal Portfolio with Transaction Costs

The aim of this paper is to investigate the impact of rebalancing frequency and transaction costs on the log-optimal portfolio, which is a portfolio that maximizes the expected logarithmic growth rate of an investor's wealth. We prove that the frequency-dependent log-optimal portfolio problem with costs is equivalent to a concave program and provide a version of the dominance theorem with costs to determine when an investor should invest all available funds in a particular asset. Then, we show that transaction costs may cause a bankruptcy issue for the frequency-dependent log-optimal portfolio. To address this issue, we approximate the problem to obtain a quadratic concave program and derive necessary and sufficient optimality conditions. Additionally, we prove a version of the two-fund theorem, which states that any convex combination of two optimal weights from the optimality conditions is still optimal. We test our proposed methods using both intraday and daily price data. Finally, we extend our empirical studies to an online trading scenario by implementing a sliding window approach. This approach enables us to solve a sequence of concave programs rather than a potentially computational complex stochastic dynamic programming problem.

false

true

false

339,585

2111.13129

Robot Skill Adaptation via Soft Actor-Critic Gaussian Mixture Models

A core challenge for an autonomous agent acting in the real world is to adapt its repertoire of skills to cope with its noisy perception and dynamics. To scale learning of skills to long-horizon tasks, robots should be able to learn and later refine their skills in a structured manner through trajectories rather than making instantaneous decisions individually at each time step. To this end, we propose the Soft Actor-Critic Gaussian Mixture Model (SAC-GMM), a novel hybrid approach that learns robot skills through a dynamical system and adapts the learned skills in their own trajectory distribution space through interactions with the environment. Our approach combines classical robotics techniques of learning from demonstration with the deep reinforcement learning framework and exploits their complementary nature. We show that our method utilizes sensors solely available during the execution of preliminarily learned skills to extract relevant features that lead to faster skill refinement. Extensive evaluations in both simulation and real-world environments demonstrate the effectiveness of our method in refining robot skills by leveraging physical interactions, high-dimensional sensory data, and sparse task completion rewards. Videos, code, and pre-trained models are available at http://sac-gmm.cs.uni-freiburg.de.

false

true

false

true

false

268,195

2012.12336

An Approach to Deploy Interactive Robotic Simulators on the Web for HRI Experiments: Results in Social Robot Navigation

Evaluation of social robot navigation inherently requires human input due to its qualitative nature. Motivated by the need to scale human evaluation, we propose a general method for deploying interactive, rich-client robotic simulations on the web. Prior approaches implement specific web-compatible simulators or provide tools to build a simulator for a specific study. Instead, our approach builds on standard Linux tools to share a graphical desktop with remote users. We leverage these tools to deploy simulators on the web that would typically be constrained to desktop computing environments. As an example implementation of our approach, we introduce the SEAN Experimental Platform (SEAN-EP). With SEAN-EP, remote users can virtually interact with a mobile robot in the Social Environment for Autonomous Navigation, without installing any software on their computer or needing specialized hardware. We validated that SEAN-EP could quickly scale the collection of human feedback and its usability through an online survey. In addition, we compared human feedback from participants that interacted with a robot using SEAN-EP with feedback obtained through a more traditional video survey. Our results suggest that human perceptions of robots may differ based on whether they interact with the robots in simulation or observe them in videos. Also, they suggest that people perceive the surveys with interactive simulations as less mentally demanding than video surveys.

false

true

false

212,893

1909.05145

Appearance invariant Entry-Exit matching using visual soft biometric traits

The problem of appearance invariant subject recognition for Entry-Exit surveillance applications is addressed. A novel Semantic Entry-Exit matching model that makes use of ancillary information about subjects such as height, build, complexion and clothing color to endorse exit of every subject who had entered private area is proposed in this paper. The proposed method is robust to variations in clothing. Each describing attribute is given equal weight while computing the matching score and hence the proposed model achieves high rank-k accuracy on benchmark datasets. The soft biometric traits used as a combination though cannot achieve high rank-1 accuracy, it helps to narrow down the search to match using reliable biometric traits such as gait and face whose learning and matching time is costlier when compared to the visual soft biometrics.

false

true

false

145,010

2501.05700

Linguistic Entity Masking to Improve Cross-Lingual Representation of Multilingual Language Models for Low-Resource Languages

Multilingual Pre-trained Language models (multiPLMs), trained on the Masked Language Modelling (MLM) objective are commonly being used for cross-lingual tasks such as bitext mining. However, the performance of these models is still suboptimal for low-resource languages (LRLs). To improve the language representation of a given multiPLM, it is possible to further pre-train it. This is known as continual pre-training. Previous research has shown that continual pre-training with MLM and subsequently with Translation Language Modelling (TLM) improves the cross-lingual representation of multiPLMs. However, during masking, both MLM and TLM give equal weight to all tokens in the input sequence, irrespective of the linguistic properties of the tokens. In this paper, we introduce a novel masking strategy, Linguistic Entity Masking (LEM) to be used in the continual pre-training step to further improve the cross-lingual representations of existing multiPLMs. In contrast to MLM and TLM, LEM limits masking to the linguistic entity types nouns, verbs and named entities, which hold a higher prominence in a sentence. Secondly, we limit masking to a single token within the linguistic entity span thus keeping more context, whereas, in MLM and TLM, tokens are masked randomly. We evaluate the effectiveness of LEM using three downstream tasks, namely bitext mining, parallel data curation and code-mixed sentiment analysis using three low-resource language pairs English-Sinhala, English-Tamil, and Sinhala-Tamil. Experiment results show that continually pre-training a multiPLM with LEM outperforms a multiPLM continually pre-trained with MLM+TLM for all three tasks.

false

true

false

523,687

2409.14907

Knowledge Planning in Large Language Models for Domain-Aligned Counseling Summarization

In mental health counseling, condensing dialogues into concise and relevant summaries (aka counseling notes) holds pivotal significance. Large Language Models (LLMs) exhibit remarkable capabilities in various generative tasks; however, their adaptation to domain-specific intricacies remains challenging, especially within mental health contexts. Unlike standard LLMs, mental health experts first plan to apply domain knowledge in writing summaries. Our work enhances LLMs' ability by introducing a novel planning engine to orchestrate structuring knowledge alignment. To achieve high-order planning, we divide knowledge encapsulation into two major phases: (i) holding dialogue structure and (ii) incorporating domain-specific knowledge. We employ a planning engine on Llama-2, resulting in a novel framework, PIECE. Our proposed system employs knowledge filtering-cum-scaffolding to encapsulate domain knowledge. Additionally, PIECE leverages sheaf convolution learning to enhance its understanding of the dialogue's structural nuances. We compare PIECE with 14 baseline methods and observe a significant improvement across ROUGE and Bleurt scores. Further, expert evaluation and analyses validate the generation quality to be effective, sometimes even surpassing the gold standard. We further benchmark PIECE with other LLMs and report improvement, including Llama-2 (+2.72%), Mistral (+2.04%), and Zephyr (+1.59%), to justify the generalizability of the planning engine.

false

true

false

490,684

1912.11853

Domain Adaptation Regularization for Spectral Pruning

Deep Neural Networks (DNNs) have recently been achieving state-of-the-art performance on a variety of computer vision related tasks. However, their computational cost limits their ability to be implemented in embedded systems with restricted resources or strict latency constraints. Model compression has therefore been an active field of research to overcome this issue. Additionally, DNNs typically require massive amounts of labeled data to be trained. This represents a second limitation to their deployment. Domain Adaptation (DA) addresses this issue by allowing knowledge learned on one labeled source distribution to be transferred to a target distribution, possibly unlabeled. In this paper, we investigate on possible improvements of compression methods in DA setting. We focus on a compression method that was previously developed in the context of a single data distribution and show that, with a careful choice of data to use during compression and additional regularization terms directly related to DA objectives, it is possible to improve compression results. We also show that our method outperforms an existing compression method studied in the DA setting by a large margin for high compression rates. Although our work is based on one specific compression method, we also outline some general guidelines for improving compression in DA setting.

false

true

false

true

false

158,683

1710.00756

Progressive Color Transfer with Dense Semantic Correspondences

We propose a new algorithm for color transfer between images that have perceptually similar semantic structures. We aim to achieve a more accurate color transfer that leverages semantically-meaningful dense correspondence between images. To accomplish this, our algorithm uses neural representations for matching. Additionally, the color transfer should be spatially variant and globally coherent. Therefore, our algorithm optimizes a local linear model for color transfer satisfying both local and global constraints. Our proposed approach jointly optimizes matching and color transfer, adopting a coarse-to-fine strategy. The proposed method can be successfully extended from one-to-one to one-to-many color transfer. The latter further addresses the problem of mismatching elements of the input image. We validate our proposed method by testing it on a large variety of image content.

false

true

false

81,912

2109.03237

MRI Reconstruction Using Deep Energy-Based Model

Purpose: Although recent deep energy-based generative models (EBMs) have shown encouraging results in many image generation tasks, how to take advantage of the self-adversarial cogitation in deep EBMs to boost the performance of Magnetic Resonance Imaging (MRI) reconstruction is still desired. Methods: With the successful application of deep learning in a wide range of MRI reconstruction, a line of emerging research involves formulating an optimization-based reconstruction method in the space of a generative model. Leveraging this, a novel regularization strategy is introduced in this article which takes advantage of self-adversarial cogitation of the deep energy-based model. More precisely, we advocate for alternative learning a more powerful energy-based model with maximum likelihood estimation to obtain the deep energy-based information, represented as image prior. Simultaneously, implicit inference with Langevin dynamics is a unique property of re-construction. In contrast to other generative models for reconstruction, the proposed method utilizes deep energy-based information as the image prior in reconstruction to improve the quality of image. Results: Experiment results that imply the proposed technique can obtain remarkable performance in terms of high reconstruction accuracy that is competitive with state-of-the-art methods, and does not suffer from mode collapse. Conclusion: Algorithmically, an iterative approach was presented to strengthen EBM training with the gradient of energy network. The robustness and the reproducibility of the algorithm were also experimentally validated. More importantly, the proposed reconstruction framework can be generalized for most MRI reconstruction scenarios.

false

true

false

254,002

2209.07841

Findings of the Shared Task on Multilingual Coreference Resolution

This paper presents an overview of the shared task on multilingual coreference resolution associated with the CRAC 2022 workshop. Shared task participants were supposed to develop trainable systems capable of identifying mentions and clustering them according to identity coreference. The public edition of CorefUD 1.0, which contains 13 datasets for 10 languages, was used as the source of training and evaluation data. The CoNLL score used in previous coreference-oriented shared tasks was used as the main evaluation metric. There were 8 coreference prediction systems submitted by 5 participating teams; in addition, there was a competitive Transformer-based baseline system provided by the organizers at the beginning of the shared task. The winner system outperformed the baseline by 12 percentage points (in terms of the CoNLL scores averaged across all datasets for individual languages).

false

true

false

317,913