JohnVitz/NLP_Assignment_3_part_3 · Datasets at Hugging Face

1610.00321

Low-dose CT denoising with convolutional neural network

To reduce the potential radiation risk, low-dose CT has attracted much attention. However, simply lowering the radiation dose will lead to significant deterioration of the image quality. In this paper, we propose a noise reduction method for low-dose CT via deep neural network without accessing original projection data. A deep convolutional neural network is trained to transform low-dose CT images towards normal-dose CT images, patch by patch. Visual and quantitative evaluation demonstrates a competing performance of the proposed method.

false

true

false

61,820

1803.05123

Defending against Adversarial Attack towards Deep Neural Networks via Collaborative Multi-task Training

Deep neural networks (DNNs) are known to be vulnerable to adversarial examples which contain human-imperceptible perturbations. A series of defending methods, either proactive defence or reactive defence, have been proposed in the recent years. However, most of the methods can only handle specific attacks. For example, proactive defending methods are invalid against grey-box or white-box attacks, while reactive defending methods are challenged by low-distortion adversarial examples or transferring adversarial examples. This becomes a critical problem since a defender usually does not have the type of the attack as a priori knowledge. Moreover, existing two-pronged defences (e.g., MagNet), which take advantages of both proactive and reactive methods, have been reported as broken under transferring attacks. To address this problem, this paper proposed a novel defensive framework based on collaborative multi-task training, aiming at providing defence for different types of attacks. The proposed defence first encodes training labels into label pairs and counters black-box attacks leveraging adversarial training supervised by the encoded label pairs. The defence further constructs a detector to identify and reject high-confidence adversarial examples that bypass the black-box defence. In addition, the proposed collaborative architecture can prevent adversaries from finding valid adversarial examples when the defence strategy is exposed. In the experiments, we evaluated our defence against four state-of-the-art attacks on $MNIST$ and $CIFAR10$ datasets. The results showed that our defending method achieved up to $96.3\%$ classification accuracy on black-box adversarial examples, and detected up to $98.7\%$ of the high confidence adversarial examples. It only decreased the model accuracy on benign example classification by $2.1\%$ for the $CIFAR10$ dataset.

false

true

false

true

false

92,580

2209.08461

Random Fourier Features for Asymmetric Kernels

The random Fourier features (RFFs) method is a powerful and popular technique in kernel approximation for scalability of kernel methods. The theoretical foundation of RFFs is based on the Bochner theorem that relates symmetric, positive definite (PD) functions to probability measures. This condition naturally excludes asymmetric functions with a wide range applications in practice, e.g., directed graphs, conditional probability, and asymmetric kernels. Nevertheless, understanding asymmetric functions (kernels) and its scalability via RFFs is unclear both theoretically and empirically. In this paper, we introduce a complex measure with the real and imaginary parts corresponding to four finite positive measures, which expands the application scope of the Bochner theorem. By doing so, this framework allows for handling classical symmetric, PD kernels via one positive measure; symmetric, non-positive definite kernels via signed measures; and asymmetric kernels via complex measures, thereby unifying them into a general framework by RFFs, named AsK-RFFs. Such approximation scheme via complex measures enjoys theoretical guarantees in the perspective of the uniform convergence. In algorithmic implementation, to speed up the kernel approximation process, which is expensive due to the calculation of total mass, we employ a subset-based fast estimation method that optimizes total masses on a sub-training set, which enjoys computational efficiency in high dimensions. Our AsK-RFFs method is empirically validated on several typical large-scale datasets and achieves promising kernel approximation performance, which demonstrate the effectiveness of AsK-RFFs.

false

true

false

318,136

2410.09904

Equitable Access to Justice: Logical LLMs Show Promise

The costs and complexity of the American judicial system limit access to legal solutions for many Americans. Large language models (LLMs) hold great potential to improve access to justice. However, a major challenge in applying AI and LLMs in legal contexts, where consistency and reliability are crucial, is the need for System 2 reasoning. In this paper, we explore the integration of LLMs with logic programming to enhance their ability to reason, bringing their strategic capabilities closer to that of a skilled lawyer. Our objective is to translate laws and contracts into logic programs that can be applied to specific legal cases, with a focus on insurance contracts. We demonstrate that while GPT-4o fails to encode a simple health insurance contract into logical code, the recently released OpenAI o1-preview model succeeds, exemplifying how LLMs with advanced System 2 reasoning capabilities can expand access to justice.

false

true

false

true

false

true

497,822

2110.04667

Competitive Perimeter Defense of Conical Environments

We consider a perimeter defense problem in a planar conical environment in which a single vehicle, having a finite capture radius, aims to defend a concentric perimeter from mobile intruders. The intruders are arbitrarily released at the circumference of the environment and they move radially toward the perimeter with fixed speed. We present a competitive analysis approach to this problem by measuring the performance of multiple online algorithms for the vehicle against arbitrary inputs, relative to an optimal offline algorithm that has information about entire input sequence in advance. In particular, we establish two necessary conditions on the parameter space to guarantee (i) finite competitiveness of any algorithm and (ii) a competitive ratio of at least 2 for any algorithm. We then design and analyze three online algorithms and characterize parameter regimes in which they have finite competitive ratios. Specifically, our first two algorithms are provably 1, and 2-competitive, respectively, whereas our third algorithm exhibits different competitive ratios in different regimes of problem parameters. Finally, we provide a numerical plot in the parameter space to reveal additional insights into the relative performance of our algorithms.

false

true

false

true

259,991

2412.15525

Generalized Back-Stepping Experience Replay in Sparse-Reward Environments

Back-stepping experience replay (BER) is a reinforcement learning technique that can accelerate learning efficiency in reversible environments. BER trains an agent with generated back-stepping transitions of collected experiences and normal forward transitions. However, the original algorithm is designed for a dense-reward environment that does not require complex exploration, limiting the BER technique to demonstrate its full potential. Herein, we propose an enhanced version of BER called Generalized BER (GBER), which extends the original algorithm to sparse-reward environments, particularly those with complex structures that require the agent to explore. GBER improves the performance of BER by introducing relabeling mechanism and applying diverse sampling strategies. We evaluate our modified version, which is based on a goal-conditioned deep deterministic policy gradient offline learning algorithm, across various maze navigation environments. The experimental results indicate that the GBER algorithm can significantly boost the performance and stability of the baseline algorithm in various sparse-reward environments, especially those with highly structural symmetricity.

false

true

false

true

false

519,156

2408.01262

RAGEval: Scenario Specific RAG Evaluation Dataset Generation Framework

Retrieval-Augmented Generation (RAG) is a powerful approach that enables large language models (LLMs) to incorporate external knowledge. However, evaluating the effectiveness of RAG systems in specialized scenarios remains challenging due to the high costs of data construction and the lack of suitable evaluation metrics. This paper introduces RAGEval, a framework designed to assess RAG systems across diverse scenarios by generating high-quality documents, questions, answers, and references through a schema-based pipeline. With a focus on factual accuracy, we propose three novel metrics Completeness, Hallucination, and Irrelevance to rigorously evaluate LLM-generated responses. Experimental results show that RAGEval outperforms zero-shot and one-shot methods in terms of clarity, safety, conformity, and richness of generated samples. Furthermore, the use of LLMs for scoring the proposed metrics demonstrates a high level of consistency with human evaluations. RAGEval establishes a new paradigm for evaluating RAG systems in real-world applications.

false

true

false

true

false

478,161

1504.08153

Principal Patterns on Graphs: Discovering Coherent Structures in Datasets

Graphs are now ubiquitous in almost every field of research. Recently, new research areas devoted to the analysis of graphs and data associated to their vertices have emerged. Focusing on dynamical processes, we propose a fast, robust and scalable framework for retrieving and analyzing recurring patterns of activity on graphs. Our method relies on a novel type of multilayer graph that encodes the spreading or propagation of events between successive time steps. We demonstrate the versatility of our method by applying it on three different real-world examples. Firstly, we study how rumor spreads on a social network. Secondly, we reveal congestion patterns of pedestrians in a train station. Finally, we show how patterns of audio playlists can be used in a recommender system. In each example, relevant information previously hidden in the data is extracted in a very efficient manner, emphasizing the scalability of our method. With a parallel implementation scaling linearly with the size of the dataset, our framework easily handles millions of nodes on a single commodity server.

false

true

false

42,621

1901.07884

Rank consistent ordinal regression for neural networks with application to age estimation

In many real-world prediction tasks, class labels include information about the relative ordering between labels, which is not captured by commonly-used loss functions such as multi-category cross-entropy. Recently, the deep learning community adopted ordinal regression frameworks to take such ordering information into account. Neural networks were equipped with ordinal regression capabilities by transforming ordinal targets into binary classification subtasks. However, this method suffers from inconsistencies among the different binary classifiers. To resolve these inconsistencies, we propose the COnsistent RAnk Logits (CORAL) framework with strong theoretical guarantees for rank-monotonicity and consistent confidence scores. Moreover, the proposed method is architecture-agnostic and can extend arbitrary state-of-the-art deep neural network classifiers for ordinal regression tasks. The empirical evaluation of the proposed rank-consistent method on a range of face-image datasets for age prediction shows a substantial reduction of the prediction error compared to the reference ordinal regression network.

false

true

false

119,328

2407.02362

Fast, Scalable, Energy-Efficient Non-element-wise Matrix Multiplication on FPGA

Modern Neural Network (NN) architectures heavily rely on vast numbers of multiply-accumulate arithmetic operations, constituting the predominant computational cost. Therefore, this paper proposes a high-throughput, scalable and energy efficient non-element-wise matrix multiplication unit on FPGAs as a basic component of the NNs. We firstly streamline inter-layer and intra-layer redundancies of MADDNESS algorithm, a LUT-based approximate matrix multiplication, to design a fast, efficient scalable approximate matrix multiplication module termed "Approximate Multiplication Unit (AMU)". The AMU optimizes LUT-based matrix multiplications further through dedicated memory management and access design, decoupling computational overhead from input resolution and boosting FPGA-based NN accelerator efficiency significantly. The experimental results show that using our AMU achieves up to 9x higher throughput and 112x higher energy efficiency over the state-of-the-art solutions for the FPGA-based Quantised Neural Network (QNN) accelerators.

false

true

false

true

false

true

469,696

2010.06261

Neighborhood Preserving Kernels for Attributed Graphs

We describe the design of a reproducing kernel suitable for attributed graphs, in which the similarity between the two graphs is defined based on the neighborhood information of the graph nodes with the aid of a product graph formulation. We represent the proposed kernel as the weighted sum of two other kernels of which one is an R-convolution kernel that processes the attribute information of the graph and the other is an optimal assignment kernel that processes label information. They are formulated in such a way that the edges processed as part of the kernel computation have the same neighborhood properties and hence the kernel proposed makes a well-defined correspondence between regions processed in graphs. These concepts are also extended to the case of the shortest paths. We identified the state-of-the-art kernels that can be mapped to such a neighborhood preserving framework. We found that the kernel value of the argument graphs in each iteration of the Weisfeiler-Lehman color refinement algorithm can be obtained recursively from the product graph formulated in our method. By incorporating the proposed kernel on support vector machines we analyzed the real-world data sets and it has shown superior performance in comparison with that of the other state-of-the-art graph kernels.

false

true

false

true

false

200,428

2104.02987

Plinius: Secure and Persistent Machine Learning Model Training

With the increasing popularity of cloud based machine learning (ML) techniques there comes a need for privacy and integrity guarantees for ML data. In addition, the significant scalability challenges faced by DRAM coupled with the high access-times of secondary storage represent a huge performance bottleneck for ML systems. While solutions exist to tackle the security aspect, performance remains an issue. Persistent memory (PM) is resilient to power loss (unlike DRAM), provides fast and fine-granular access to memory (unlike disk storage) and has latency and bandwidth close to DRAM (in the order of ns and GB/s, respectively). We present PLINIUS, a ML framework using Intel SGX enclaves for secure training of ML models and PM for fault tolerance guarantees. P LINIUS uses a novel mirroring mechanism to create and maintain (i) encrypted mirror copies of ML models on PM, and (ii) encrypted training data in byte-addressable PM, for near-instantaneous data recovery after a system failure. Compared to disk-based checkpointing systems,PLINIUS is 3.2x and 3.7x faster respectively for saving and restoring models on real PM hardware, achieving robust and secure ML model training in SGX enclaves.

false

true

false

true

false

228,930

2312.17345

3VL: Using Trees to Improve Vision-Language Models' Interpretability

Vision-Language models (VLMs) have proven to be effective at aligning image and text representations, producing superior zero-shot results when transferred to many downstream tasks. However, these representations suffer from some key shortcomings in understanding Compositional Language Concepts (CLC), such as recognizing objects' attributes, states, and relations between different objects. Moreover, VLMs typically have poor interpretability, making it challenging to debug and mitigate compositional-understanding failures. In this work, we introduce the architecture and training technique of Tree-augmented Vision-Language (3VL) model accompanied by our proposed Anchor inference method and Differential Relevance (DiRe) interpretability tool. By expanding the text of an arbitrary image-text pair into a hierarchical tree structure using language analysis tools, 3VL allows the induction of this structure into the visual representation learned by the model, enhancing its interpretability and compositional reasoning. Additionally, we show how Anchor, a simple technique for text unification, can be used to filter nuisance factors while increasing CLC understanding performance, e.g., on the fundamental VL-Checklist benchmark. We also show how DiRe, which performs a differential comparison between VLM relevancy maps, enables us to generate compelling visualizations of the reasons for a model's success or failure. Our code is available at: https://github.com/niryellinek/3VL.

false

true

false

418,720

2409.19912

HYDRA-FL: Hybrid Knowledge Distillation for Robust and Accurate Federated Learning

Data heterogeneity among Federated Learning (FL) users poses a significant challenge, resulting in reduced global model performance. The community has designed various techniques to tackle this issue, among which Knowledge Distillation (KD)-based techniques are common. While these techniques effectively improve performance under high heterogeneity, they inadvertently cause higher accuracy degradation under model poisoning attacks (known as attack amplification). This paper presents a case study to reveal this critical vulnerability in KD-based FL systems. We show why KD causes this issue through empirical evidence and use it as motivation to design a hybrid distillation technique. We introduce a novel algorithm, Hybrid Knowledge Distillation for Robust and Accurate FL (HYDRA-FL), which reduces the impact of attacks in attack scenarios by offloading some of the KD loss to a shallow layer via an auxiliary classifier. We model HYDRA-FL as a generic framework and adapt it to two KD-based FL algorithms, FedNTD and MOON. Using these two as case studies, we demonstrate that our technique outperforms baselines in attack settings while maintaining comparable performance in benign settings.

false

true

false

true

false

492,905

2111.14690

DanceTrack: Multi-Object Tracking in Uniform Appearance and Diverse Motion

A typical pipeline for multi-object tracking (MOT) is to use a detector for object localization, and following re-identification (re-ID) for object association. This pipeline is partially motivated by recent progress in both object detection and re-ID, and partially motivated by biases in existing tracking datasets, where most objects tend to have distinguishing appearance and re-ID models are sufficient for establishing associations. In response to such bias, we would like to re-emphasize that methods for multi-object tracking should also work when object appearance is not sufficiently discriminative. To this end, we propose a large-scale dataset for multi-human tracking, where humans have similar appearance, diverse motion and extreme articulation. As the dataset contains mostly group dancing videos, we name it "DanceTrack". We expect DanceTrack to provide a better platform to develop more MOT algorithms that rely less on visual discrimination and depend more on motion analysis. We benchmark several state-of-the-art trackers on our dataset and observe a significant performance drop on DanceTrack when compared against existing benchmarks. The dataset, project code and competition server are released at: \url{https://github.com/DanceTrack}.

false

true

false

268,673

2305.14709

Regret Matching+: (In)Stability and Fast Convergence in Games

Regret Matching+ (RM+) and its variants are important algorithms for solving large-scale games. However, a theoretical understanding of their success in practice is still a mystery. Moreover, recent advances on fast convergence in games are limited to no-regret algorithms such as online mirror descent, which satisfy stability. In this paper, we first give counterexamples showing that RM+ and its predictive version can be unstable, which might cause other players to suffer large regret. We then provide two fixes: restarting and chopping off the positive orthant that RM+ works in. We show that these fixes are sufficient to get $O(T^{1/4})$ individual regret and $O(1)$ social regret in normal-form games via RM+ with predictions. We also apply our stabilizing techniques to clairvoyant updates in the uncoupled learning setting for RM+ and prove desirable results akin to recent works for Clairvoyant online mirror descent. Our experiments show the advantages of our algorithms over vanilla RM+-based algorithms in matrix and extensive-form games.

false

true

false

true

367,201

2310.19232

Adapter Pruning using Tropical Characterization

Adapters are widely popular parameter-efficient transfer learning approaches in natural language processing that insert trainable modules in between layers of a pre-trained language model. Apart from several heuristics, however, there has been a lack of studies analyzing the optimal number of adapter parameters needed for downstream applications. In this paper, we propose an adapter pruning approach by studying the tropical characteristics of trainable modules. We cast it as an optimization problem that aims to prune parameters from the adapter layers without changing the orientation of underlying tropical hypersurfaces. Our experiments on five NLP datasets show that tropical geometry tends to identify more relevant parameters to prune when compared with the magnitude-based baseline, while a combined approach works best across the tasks.

false

true

false

403,909

1912.12096

Coverage Analysis of Relay Assisted Millimeter Wave Cellular Networks with Spatial Correlation

We propose a novel analytical framework for evaluating the coverage performance of a millimeter wave (mmWave) cellular network where idle user equipments (UEs) act as relays. In this network, the base station (BS) adopts either the direct mode to transmit to the destination UE, or the relay mode if the direct mode fails, where the BS transmits to the relay UE and then the relay UE transmits to the destination UE. To address the drastic rotational movements of destination UEs in practice, we propose to adopt selection combining at destination UEs. New expression is derived for the signal-to-interference-plus-noise ratio (SINR) coverage probability of the network. Using numerical results, we first demonstrate the accuracy of our new expression. Then we show that ignoring spatial correlation, which has been commonly adopted in the literature, leads to severe overestimation of the SINR coverage probability. Furthermore, we show that introducing relays into a mmWave cellular network vastly improves the coverage performance. In addition, we show that the optimal BS density maximizing the SINR coverage probability can be determined by using our analysis.

false

true

false

158,751

2310.18877

Pre-trained Speech Processing Models Contain Human-Like Biases that Propagate to Speech Emotion Recognition

Previous work has established that a person's demographics and speech style affect how well speech processing models perform for them. But where does this bias come from? In this work, we present the Speech Embedding Association Test (SpEAT), a method for detecting bias in one type of model used for many speech tasks: pre-trained models. The SpEAT is inspired by word embedding association tests in natural language processing, which quantify intrinsic bias in a model's representations of different concepts, such as race or valence (something's pleasantness or unpleasantness) and capture the extent to which a model trained on large-scale socio-cultural data has learned human-like biases. Using the SpEAT, we test for six types of bias in 16 English speech models (including 4 models also trained on multilingual data), which come from the wav2vec 2.0, HuBERT, WavLM, and Whisper model families. We find that 14 or more models reveal positive valence (pleasantness) associations with abled people over disabled people, with European-Americans over African-Americans, with females over males, with U.S. accented speakers over non-U.S. accented speakers, and with younger people over older people. Beyond establishing that pre-trained speech models contain these biases, we also show that they can have real world effects. We compare biases found in pre-trained models to biases in downstream models adapted to the task of Speech Emotion Recognition (SER) and find that in 66 of the 96 tests performed (69%), the group that is more associated with positive valence as indicated by the SpEAT also tends to be predicted as speaking with higher valence by the downstream model. Our work provides evidence that, like text and image-based models, pre-trained speech based-models frequently learn human-like biases. Our work also shows that bias found in pre-trained models can propagate to the downstream task of SER.

false

true

false

true

false

true

false

403,743

2406.13232

Towards Robust Evaluation: A Comprehensive Taxonomy of Datasets and Metrics for Open Domain Question Answering in the Era of Large Language Models

Open Domain Question Answering (ODQA) within natural language processing involves building systems that answer factual questions using large-scale knowledge corpora. Recent advances stem from the confluence of several factors, such as large-scale training datasets, deep learning techniques, and the rise of large language models. High-quality datasets are used to train models on realistic scenarios and enable the evaluation of the system on potentially unseen data. Standardized metrics facilitate comparisons between different ODQA systems, allowing researchers to objectively track advancements in the field. Our study presents a thorough examination of the current landscape of ODQA benchmarking by reviewing 52 datasets and 20 evaluation techniques across textual and multimodal modalities. We introduce a novel taxonomy for ODQA datasets that incorporates both the modality and difficulty of the question types. Additionally, we present a structured organization of ODQA evaluation metrics along with a critical analysis of their inherent trade-offs. Our study aims to empower researchers by providing a framework for the robust evaluation of modern question-answering systems. We conclude by identifying the current challenges and outlining promising avenues for future research and development.

false

true

false

true

false

465,762

1307.6883

A gradient descent technique coupled with a dynamic simulation to determine the near optimum orientation of floor plan designs

A prototype tool to assist architects during the early design stage of floor plans has been developed, consisting of an Evolutionary Program for the Space Allocation Problem (EPSAP), which generates sets of floor plan alternatives according to the architect's preferences; and a Floor Plan Performance Optimization Program (FPOP), which optimizes the selected solutions according to thermal performance criteria. The design variables subject to optimization are window position and size, overhangs, fins, wall positioning, and building orientation. A procedure using a transformation operator with gradient descent, such as behavior, coupled with a dynamic simulation engine was developed for the thermal evaluation and optimization process. However, the need to evaluate all possible alternatives regarding designing variables being used during the optimization process leads to an intensive use of thermal simulation, which dramatically increases the simulation time, rendering it unpractical. An alternative approach is a smart optimization approach, which utilizes an oriented and adaptive search technique to efficiently find the near optimum solution. This paper presents the search methodology for the building orientation of floor plan designs, and the corresponding efficiency and effectiveness indicators. The calculations are based on 100 floor plan designs generated by EPSAP. All floor plans have the same design program, location, and weather data, changing only their geometry. Dynamic simulation of buildings was effectively used together with the optimization procedure in this approach to significantly improve the designs. The use of the orientation variable has been included in the algorithm.

false

true

false

true

false

26,051

2310.06218

SUBP: Soft Uniform Block Pruning for 1xN Sparse CNNs Multithreading Acceleration

The study of sparsity in Convolutional Neural Networks (CNNs) has become widespread to compress and accelerate models in environments with limited resources. By constraining N consecutive weights along the output channel to be group-wise non-zero, the recent network with 1$\times$N sparsity has received tremendous popularity for its three outstanding advantages: 1) A large amount of storage space saving by a \emph{Block Sparse Row} matrix. 2) Excellent performance at a high sparsity. 3) Significant speedups on CPUs with Advanced Vector Extensions. Recent work requires selecting and fine-tuning 1$\times$N sparse weights based on dense pre-trained weights, leading to the problems such as expensive training cost and memory access, sub-optimal model quality, as well as unbalanced workload across threads (different sparsity across output channels). To overcome them, this paper proposes a novel \emph{\textbf{S}oft \textbf{U}niform \textbf{B}lock \textbf{P}runing} (SUBP) approach to train a uniform 1$\times$N sparse structured network from scratch. Specifically, our approach tends to repeatedly allow pruned blocks to regrow to the network based on block angular redundancy and importance sampling in a uniform manner throughout the training process. It not only makes the model less dependent on pre-training, reduces the model redundancy and the risk of pruning the important blocks permanently but also achieves balanced workload. Empirically, on ImageNet, comprehensive experiments across various CNN architectures show that our SUBP consistently outperforms existing 1$\times$N and structured sparsity methods based on pre-trained models or training from scratch. Source codes and models are available at \url{https://github.com/JingyangXiang/SUBP}.

false

true

false

true

false

398,474

1708.04765

Dialogue Act Segmentation for Vietnamese Human-Human Conversational Texts

Dialog act identification plays an important role in understanding conversations. It has been widely applied in many fields such as dialogue systems, automatic machine translation, automatic speech recognition, and especially useful in systems with human-computer natural language dialogue interfaces such as virtual assistants and chatbots. The first step of identifying dialog act is identifying the boundary of the dialog act in utterances. In this paper, we focus on segmenting the utterance according to the dialog act boundaries, i.e. functional segments identification, for Vietnamese utterances. We investigate carefully functional segment identification in two approaches: (1) machine learning approach using maximum entropy (ME) and conditional random fields (CRFs); (2) deep learning approach using bidirectional Long Short-Term Memory (LSTM) with a CRF layer (Bi-LSTM-CRF) on two different conversational datasets: (1) Facebook messages (Message data); (2) transcription from phone conversations (Phone data). To the best of our knowledge, this is the first work that applies deep learning based approach to dialog act segmentation. As the results show, deep learning approach performs appreciably better as to compare with traditional machine learning approaches. Moreover, it is also the first study that tackles dialog act and functional segment identification for Vietnamese.

false

true

false

79,017

1109.2048

An Expressive Language and Efficient Execution System for Software Agents

Software agents can be used to automate many of the tedious, time-consuming information processing tasks that humans currently have to complete manually. However, to do so, agent plans must be capable of representing the myriad of actions and control flows required to perform those tasks. In addition, since these tasks can require integrating multiple sources of remote information ? typically, a slow, I/O-bound process ? it is desirable to make execution as efficient as possible. To address both of these needs, we present a flexible software agent plan language and a highly parallel execution system that enable the efficient execution of expressive agent plans. The plan language allows complex tasks to be more easily expressed by providing a variety of operators for flexibly processing the data as well as supporting subplans (for modularity) and recursion (for indeterminate looping). The executor is based on a streaming dataflow model of execution to maximize the amount of operator and data parallelism possible at runtime. We have implemented both the language and executor in a system called THESEUS. Our results from testing THESEUS show that streaming dataflow execution can yield significant speedups over both traditional serial (von Neumann) as well as non-streaming dataflow-style execution that existing software and robot agent execution systems currently support. In addition, we show how plans written in the language we present can represent certain types of subtasks that cannot be accomplished using the languages supported by network query engines. Finally, we demonstrate that the increased expressivity of our plan language does not hamper performance; specifically, we show how data can be integrated from multiple remote sources just as efficiently using our architecture as is possible with a state-of-the-art streaming-dataflow network query engine.

false

true

false

12,074

1702.07942

BARCHAN: Blob Alignment for Robust CHromatographic ANalysis

Comprehensive Two dimensional gas chromatography (GCxGC) plays a central role into the elucidation of complex samples. The automation of the identification of peak areas is of prime interest to obtain a fast and repeatable analysis of chromatograms. To determine the concentration of compounds or pseudo-compounds, templates of blobs are defined and superimposed on a reference chromatogram. The templates then need to be modified when different chromatograms are recorded. In this study, we present a chromatogram and template alignment method based on peak registration called BARCHAN. Peaks are identified using a robust mathematical morphology tool. The alignment is performed by a probabilistic estimation of a rigid transformation along the first dimension, and a non-rigid transformation in the second dimension, taking into account noise, outliers and missing peaks in a fully automated way. Resulting aligned chromatograms and masks are presented on two datasets. The proposed algorithm proves to be fast and reliable. It significantly reduces the time to results for GCxGC analysis.

false

true

false

68,871

2402.05126

Graph Neural Network and NER-Based Text Summarization

With the abundance of data and information in todays time, it is nearly impossible for man, or, even machine, to go through all of the data line by line. What one usually does is to try to skim through the lines and retain the absolutely important information, that in a more formal term is called summarization. Text summarization is an important task that aims to compress lengthy documents or articles into shorter, coherent representations while preserving the core information and meaning. This project introduces an innovative approach to text summarization, leveraging the capabilities of Graph Neural Networks (GNNs) and Named Entity Recognition (NER) systems. GNNs, with their exceptional ability to capture and process the relational data inherent in textual information, are adept at understanding the complex structures within large documents. Meanwhile, NER systems contribute by identifying and emphasizing key entities, ensuring that the summarization process maintains a focus on the most critical aspects of the text. By integrating these two technologies, our method aims to enhances the efficiency of summarization and also tries to ensures a high degree relevance in the condensed content. This project, therefore, offers a promising direction for handling the ever increasing volume of textual data in an information-saturated world.

false

true

false

true

false

427,734

1802.09058

Seeing Small Faces from Robust Anchor's Perspective

This paper introduces a novel anchor design to support anchor-based face detection for superior scale-invariant performance, especially on tiny faces. To achieve this, we explicitly address the problem that anchor-based detectors drop performance drastically on faces with tiny sizes, e.g. less than 16x16 pixels. In this paper, we investigate why this is the case. We discover that current anchor design cannot guarantee high overlaps between tiny faces and anchor boxes, which increases the difficulty of training. The new Expected Max Overlapping (EMO) score is proposed which can theoretically explain the low overlapping issue and inspire several effective strategies of new anchor design leading to higher face overlaps, including anchor stride reduction with new network architectures, extra shifted anchors, and stochastic face shifting. Comprehensive experiments show that our proposed method significantly outperforms the baseline anchor-based detector, while consistently achieving state-of-the-art results on challenging face detection datasets with competitive runtime speed.

false

true

false

91,254

2109.03814

Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers

Panoptic segmentation involves a combination of joint semantic segmentation and instance segmentation, where image contents are divided into two types: things and stuff. We present Panoptic SegFormer, a general framework for panoptic segmentation with transformers. It contains three innovative components: an efficient deeply-supervised mask decoder, a query decoupling strategy, and an improved post-processing method. We also use Deformable DETR to efficiently process multi-scale features, which is a fast and efficient version of DETR. Specifically, we supervise the attention modules in the mask decoder in a layer-wise manner. This deep supervision strategy lets the attention modules quickly focus on meaningful semantic regions. It improves performance and reduces the number of required training epochs by half compared to Deformable DETR. Our query decoupling strategy decouples the responsibilities of the query set and avoids mutual interference between things and stuff. In addition, our post-processing strategy improves performance without additional costs by jointly considering classification and segmentation qualities to resolve conflicting mask overlaps. Our approach increases the accuracy 6.2\% PQ over the baseline DETR model. Panoptic SegFormer achieves state-of-the-art results on COCO test-dev with 56.2\% PQ. It also shows stronger zero-shot robustness over existing methods. The code is released at \url{https://github.com/zhiqi-li/Panoptic-SegFormer}.

false

true

false

254,180

2501.06740

Rice Leaf Disease Detection: A Comparative Study Between CNN, Transformer and Non-neural Network Architectures

In nations such as Bangladesh, agriculture plays a vital role in providing livelihoods for a significant portion of the population. Identifying and classifying plant diseases early is critical to prevent their spread and minimize their impact on crop yield and quality. Various computer vision techniques can be used for such detection and classification. While CNNs have been dominant on such image classification tasks, vision transformers has become equally good in recent time also. In this paper we study the various computer vision techniques for Bangladeshi rice leaf disease detection. We use the Dhan-Shomadhan -- a Bangladeshi rice leaf disease dataset, to experiment with various CNN and ViT models. We also compared the performance of such deep neural network architecture with traditional machine learning architecture like Support Vector Machine(SVM). We leveraged transfer learning for better generalization with lower amount of training data. Among the models tested, ResNet50 exhibited the best performance over other CNN and transformer-based models making it the optimal choice for this task.

false

true

false

524,107

2203.11997

Federated Self-Supervised Learning for Acoustic Event Classification

Standard acoustic event classification (AEC) solutions require large-scale collection of data from client devices for model optimization. Federated learning (FL) is a compelling framework that decouples data collection and model training to enhance customer privacy. In this work, we investigate the feasibility of applying FL to improve AEC performance while no customer data can be directly uploaded to the server. We assume no pseudo labels can be inferred from on-device user inputs, aligning with the typical use cases of AEC. We adapt self-supervised learning to the FL framework for on-device continual learning of representations, and it results in improved performance of the downstream AEC classifiers without labeled/pseudo-labeled data available. Compared to the baseline w/o FL, the proposed method improves precision up to 20.3\% relatively while maintaining the recall. Our work differs from prior work in FL in that our approach does not require user-generated learning targets, and the data we use is collected from our Beta program and is de-identified, to maximally simulate the production settings.

false

true

false

true

false

287,105

2103.15947

Federated Learning with Taskonomy for Non-IID Data

Classical federated learning approaches incur significant performance degradation in the presence of non-IID client data. A possible direction to address this issue is forming clusters of clients with roughly IID data. Most solutions following this direction are iterative and relatively slow, also prone to convergence issues in discovering underlying cluster formations. We introduce federated learning with taskonomy (FLT) that generalizes this direction by learning the task-relatedness between clients for more efficient federated aggregation of heterogeneous data. In a one-off process, the server provides the clients with a pretrained (and fine-tunable) encoder to compress their data into a latent representation, and transmit the signature of their data back to the server. The server then learns the task-relatedness among clients via manifold learning, and performs a generalization of federated averaging. FLT can flexibly handle a generic client relatedness graph, when there are no explicit clusters of clients, as well as efficiently decompose it into (disjoint) clusters for clustered federated learning. We demonstrate that FLT not only outperforms the existing state-of-the-art baselines in non-IID scenarios but also offers improved fairness across clients.

false

true

false

true

227,385

2310.00906

A Decentralized Cooperative Navigation Approach for Visual Homing Networks

Visual homing is a lightweight approach to visual navigation. Given the stored information of an initial 'home' location, the navigation task back to this location is achieved from any other location by comparing the stored home information to the current image and extracting a motion vector. A challenge that constrains the applicability of visual homing is that the home location must be within the robot's field of view to initiate the homing process. Thus, we propose a blockchain approach to visual navigation for a heterogeneous robot team over a wide area of visual navigation. Because it does not require map data structures, the approach is useful for robot platforms with a small computational footprint, and because it leverages current visual information, it supports a resilient and adaptive path selection. Further, we present a lightweight Proof-of-Work (PoW) mechanism for reaching consensus in the untrustworthy visual homing network.

false

true

false

true

false

true

396,203

2110.13670

W-Net: A Two-Stage Convolutional Network for Nucleus Detection in Histopathology Image

Pathological diagnosis is the gold standard for cancer diagnosis, but it is labor-intensive, in which tasks such as cell detection, classification, and counting are particularly prominent. A common solution for automating these tasks is using nucleus segmentation technology. However, it is hard to train a robust nucleus segmentation model, due to several challenging problems, the nucleus adhesion, stacking, and excessive fusion with the background. Recently, some researchers proposed a series of automatic nucleus segmentation methods based on point annotation, which can significant improve the model performance. Nevertheless, the point annotation needs to be marked by experienced pathologists. In order to take advantage of segmentation methods based on point annotation, further alleviate the manual workload, and make cancer diagnosis more efficient and accurate, it is necessary to develop an automatic nucleus detection algorithm, which can automatically and efficiently locate the position of the nucleus in the pathological image and extract valuable information for pathologists. In this paper, we propose a W-shaped network for automatic nucleus detection. Different from the traditional U-Net based method, mapping the original pathology image to the target mask directly, our proposed method split the detection task into two sub-tasks. The first sub-task maps the original pathology image to the binary mask, then the binary mask is mapped to the density mask in the second sub-task. After the task is split, the task's difficulty is significantly reduced, and the network's overall performance is improved.

false

true

false

263,275

1506.00982

Game Theory for Signal Processing in Networks

In this tutorial, the basics of game theory are introduced along with an overview of its most recent and emerging applications in signal processing. One of the main features of this contribution is to gather in a single paper some fundamental game-theoretic notions and tools which, over the past few years, have become widely spread over a large number of papers. In particular, both strategic-form and coalition-form games are described in details while the key connections and differences between them are outlined. Moreover, a particular attention is also devoted to clarify the connections between strategic-form games and distributed optimization and learning algorithms. Beyond an introduction to the basic concepts and main solution approaches, several carefully designed examples are provided to allow a better understanding of how to apply the described tools.

false

true

false

true

43,737

2003.00187

Augmented Cyclic Consistency Regularization for Unpaired Image-to-Image Translation

Unpaired image-to-image (I2I) translation has received considerable attention in pattern recognition and computer vision because of recent advancements in generative adversarial networks (GANs). However, due to the lack of explicit supervision, unpaired I2I models often fail to generate realistic images, especially in challenging datasets with different backgrounds and poses. Hence, stabilization is indispensable for GANs and applications of I2I translation. Herein, we propose Augmented Cyclic Consistency Regularization (ACCR), a novel regularization method for unpaired I2I translation. Our main idea is to enforce consistency regularization originating from semi-supervised learning on the discriminators leveraging real, fake, reconstructed, and augmented samples. We regularize the discriminators to output similar predictions when fed pairs of original and perturbed images. We qualitatively clarify why consistency regularization on fake and reconstructed samples works well. Quantitatively, our method outperforms the consistency regularized GAN (CR-GAN) in real-world translations and demonstrates efficacy against several data augmentation variants and cycle-consistent constraints.

false

true

false

true

false

166,217

1802.09338

An Energy Balance Based Method for Parameter Identification of a Free-Flying Robot Grasping An Unknown Object

The estimation of inertial parameters of a robotic system is crucial for better trajectory tracking performance, specially when model-based controllers are used for carrying out precise tasks. In this paper, we consider the scenario of grasping an object of unknown properties by a free-flyer space robot with limited actuation. The problem is to find the inertial parameters of the complete system after grasping has been performed. Excitation is provided in inertial space, and the excitation trajectories are found by optimization. Truncated Fourier series are used to represent the reference as well as tracked trajectory. An approach based on the energy balance between the actuation work and the rate of change of kinetic energy is introduced to calculate the number of harmonics in the Fourier series used to represent the executed trajectory, while trying to find a balance between accounting for saturation effects and keeping out noise. The effect of input saturation on parameter estimation is also studied. Simulation results using the Space CoBot free-flyer robot are presented to show the feasibility of the approach.

false

true

false

91,309

2204.07096

Detection of Degraded Acacia tree species using deep neural networks on uav drone imagery

Deep-learning-based image classification and object detection has been applied successfully to tree monitoring. However, studies of tree crowns and fallen trees, especially on flood inundated areas, remain largely unexplored. Detection of degraded tree trunks on natural environments such as water, mudflats, and natural vegetated areas is challenging due to the mixed colour image backgrounds. In this paper, Unmanned Aerial Vehicles (UAVs), or drones, with embedded RGB cameras were used to capture the fallen Acacia Xanthophloea trees from six designated plots around Lake Nakuru, Kenya. Motivated by the need to detect fallen trees around the lake, two well-established deep neural networks, i.e. Faster Region-based Convolution Neural Network (Faster R-CNN) and Retina-Net were used for fallen tree detection. A total of 7,590 annotations of three classes on 256 x 256 image patches were used for this study. Experimental results show the relevance of deep learning in this context, with Retina-Net model achieving 38.9% precision and 57.9% recall.

false

true

false

291,562

2409.15904

Unimotion: Unifying 3D Human Motion Synthesis and Understanding

We introduce Unimotion, the first unified multi-task human motion model capable of both flexible motion control and frame-level motion understanding. While existing works control avatar motion with global text conditioning, or with fine-grained per frame scripts, none can do both at once. In addition, none of the existing works can output frame-level text paired with the generated poses. In contrast, Unimotion allows to control motion with global text, or local frame-level text, or both at once, providing more flexible control for users. Importantly, Unimotion is the first model which by design outputs local text paired with the generated poses, allowing users to know what motion happens and when, which is necessary for a wide range of applications. We show Unimotion opens up new applications: 1.) Hierarchical control, allowing users to specify motion at different levels of detail, 2.) Obtaining motion text descriptions for existing MoCap data or YouTube videos 3.) Allowing for editability, generating motion from text, and editing the motion via text edits. Moreover, Unimotion attains state-of-the-art results for the frame-level text-to-motion task on the established HumanML3D dataset. The pre-trained model and code are available available on our project page at https://coral79.github.io/uni-motion/.

false

true

false

491,120

2305.15985

Resource Allocation in Cell-Free MU-MIMO Multicarrier System with Finite and Infinite Blocklength

The explosive growth of data results in more scarce spectrum resources. It is important to optimize the system performance under limited resources. In this paper, we investigate how to achieve weighted throughput (WTP) maximization for cell-free (CF) multiuser MIMO (MU-MIMO) multicarrier (MC) systems through resource allocation (RA), in the cases of finite blocklength (FBL) and infinite blocklength (INFBL) regimes. To ensure the quality of service (QoS) of each user, particularly for the block error rate (BLER) and latency in the FBL regime, the WTP gets maximized under the constraints of total power consumption and required QoS metrics. Since the channels vary in different subcarriers (SCs) and inter-user interference strengths, the WTP can be maximized by scheduling the best users in each time-frequency (TF) resource and advanced beamforming design, while the resources can be fully utilized. With this motivation, we propose a joint user scheduling (US) and beamforming design algorithm based on the successive convex approximation (SCA) and gene-aided (GA) algorithms, to address a mixed integer nonlinear programming (MINLP) problem. Numerical results demonstrate that the proposed RA outperforms the comparison schemes. And the CF system in our scenario is capable of achieving higher spectral efficiency than the centralized antenna systems (CAS).

false

true

false

367,876

1911.03263

Estimating States for Nonlinear Systems Using the Particle Filter

Kalman filtering has been traditionally applied in three application areas of estimation, state estimation, parameter estimation (a.k.a. model updating), and dual estimation. However, Kalman filter is often not sufficient when experimenting with highly uncertain nonlinear dynamic systems. In this study, a nonlinear estimator is developed by adopting a particle filter algorithm that takes advantage of measured signals. This approach is shown to significantly improve the ability to estimate states. To illustrate this approach, a model for a nonlinear device coupled with a hydraulic actuator plays the role of an actual plant and a nonlinear algebraic function is considered as an approximation of the nonlinear device, thus generating non-parametric and parametric uncertainties. Then we use displacement and force signals to improve the distribution of the states by resampling the set of particles. Finally, all of the states are estimated from these posterior density functions. A set of simulations considering three different noise levels demonstrates that the performance of the particle filter approach is superior to the Kalman filter, yielding substantially better performance when estimating nonlinear physical systems in the presence of modeling uncertainties.

false

true

false

152,592

2311.13471

Comparative Analysis of Linear Regression, Gaussian Elimination, and LU Decomposition for CT Real Estate Purchase Decisions

This paper presents a comprehensive evaluation of three distinct computational algorithms applied to the decision-making process of real estate purchases. Specifically, we analyze the efficacy of Linear Regression from Scikit-learn library, Gaussian Elimination with partial pivoting, and LU Decomposition in predicting the advisability of buying a house in the State of Connecticut based on a set of financial and market-related parameters. The algorithms' performances were compared using a dataset encompassing town-specific details, yearly data, interest rates, and median sale ratios. Our results demonstrate significant differences in predictive accuracy, with Linear Regression and LU Decomposition providing the most reliable recommendations and Gaussian Elimination showing limitations in stability and performance. The study's findings emphasize the importance of algorithm selection in predictive analytic and offer insights into the practical applications of computational methods in real estate investment strategies. By evaluating model efficacy through metrics such as R-squared scores and Mean Squared Error, we provide a nuanced understanding of each method's strengths and weaknesses, contributing valuable knowledge to the fields of real estate analysis and predictive modeling.

false

true

false

true

false

true

409,748

2204.09888

Fairness in Graph Mining: A Survey

Graph mining algorithms have been playing a significant role in myriad fields over the years. However, despite their promising performance on various graph analytical tasks, most of these algorithms lack fairness considerations. As a consequence, they could lead to discrimination towards certain populations when exploited in human-centered applications. Recently, algorithmic fairness has been extensively studied in graph-based applications. In contrast to algorithmic fairness on independent and identically distributed (i.i.d.) data, fairness in graph mining has exclusive backgrounds, taxonomies, and fulfilling techniques. In this survey, we provide a comprehensive and up-to-date introduction of existing literature under the context of fair graph mining. Specifically, we propose a novel taxonomy of fairness notions on graphs, which sheds light on their connections and differences. We further present an organized summary of existing techniques that promote fairness in graph mining. Finally, we summarize the widely used datasets in this emerging research field and provide insights on current research challenges and open questions, aiming at encouraging cross-breeding ideas and further advances.

false

true

false

true

false

292,597

2010.11026

A Large-Scale Analysis of IoT Firmware Version Distribution in the Wild

This paper examines the up-to-dateness of installed firmware versions of IoT devices accessible via public internet. It analyzes datasets of 1.06m devices collected from the IoT search engine Censys and maps the results against the latest firmware version each manufacturer offers. By applying the SEMMA data mining process, a fully scalable and adaptive approach is developed. This approach relies on three data artifacts: raw data from Censys, a mapping table with firmware versions and a keyword search list. The preliminary results confirm the heterogeneity of connected IoT devices. They show that manufacturer, device type and country influence the up-to-dateness of firmware. The results suggest users as a "weak link" as they do not update the firmware of their devices in a timely manner. However, the heterogeneity leads to results not showing a high reliability, yet.

false

true

false

202,095

1311.2123

Linear-Complexity Overhead-Optimized Random Linear Network Codes

Sparse random linear network coding (SRLNC) is an attractive technique proposed in the literature to reduce the decoding complexity of random linear network coding. Recognizing the fact that the existing SRLNC schemes are not efficient in terms of the required reception overhead, we consider the problem of designing overhead-optimized SRLNC schemes. To this end, we introduce a new design of SRLNC scheme that enjoys very small reception overhead while maintaining the main benefit of SRLNC, i.e., its linear encoding/decoding complexity. We also provide a mathematical framework for the asymptotic analysis and design of this class of codes based on density evolution (DE) equations. To the best of our knowledge, this work introduces the first DE analysis in the context of network coding. Our analysis method then enables us to design network codes with reception overheads in the order of a few percent. We also investigate the finite-length performance of the proposed codes and through numerical examples we show that our proposed codes have significantly lower reception overheads compared to all existing linear-complexity random linear network coding schemes.

false

true

false

28,289

1607.08481

A Nonlocal Denoising Algorithm for Manifold-Valued Images Using Second Order Statistics

Nonlocal patch-based methods, in particular the Bayes' approach of Lebrun, Buades and Morel (2013), are considered as state-of-the-art methods for denoising (color) images corrupted by white Gaussian noise of moderate variance. This paper is the first attempt to generalize this technique to manifold-valued images. Such images, for example images with phase or directional entries or with values in the manifold of symmetric positive definite matrices, are frequently encountered in real-world applications. Generalizing the normal law to manifolds is not canonical and different attempts have been considered. Here we focus on a straightforward intrinsic model and discuss the relation to other approaches for specific manifolds. We reinterpret the Bayesian approach of Lebrun et al. (2013) in terms of minimum mean squared error estimation, which motivates our definition of a corresponding estimator on the manifold. With this estimator at hand we present a nonlocal patch-based method for the restoration of manifold-valued images. Various proof of concept examples demonstrate the potential of the proposed algorithm.

false

true

false

59,166

2303.08128

ViperGPT: Visual Inference via Python Execution for Reasoning

Answering visual queries is a complex task that requires both visual processing and reasoning. End-to-end models, the dominant approach for this task, do not explicitly differentiate between the two, limiting interpretability and generalization. Learning modular programs presents a promising alternative, but has proven challenging due to the difficulty of learning both the programs and modules simultaneously. We introduce ViperGPT, a framework that leverages code-generation models to compose vision-and-language models into subroutines to produce a result for any query. ViperGPT utilizes a provided API to access the available modules, and composes them by generating Python code that is later executed. This simple approach requires no further training, and achieves state-of-the-art results across various complex visual tasks.

false

true

false

351,520

1812.10859

Signal Classification under structure sparsity constraints

Object Classification is a key direction of research in signal and image processing, computer vision and artificial intelligence. The goal is to come up with algorithms that automatically analyze images and put them in predefined categories. This dissertation focuses on the theory and application of sparse signal processing and learning algorithms for image processing and computer vision, especially object classification problems. A key emphasis of this work is to formulate novel optimization problems for learning dictionary and structured sparse representations. Tractable solutions are proposed subsequently for the corresponding optimization problems. An important goal of this dissertation is to demonstrate the wide applications of these algorithmic tools for real-world applications. To that end, we explored important problems in the areas of: 1. Medical imaging: histopathological images acquired from mammalian tissues, human breast tissues, and human brain tissues. 2. Low-frequency (UHF to L-band) ultra-wideband (UWB) synthetic aperture radar: detecting bombs and mines buried under rough surfaces. 3. General object classification: face, flowers, objects, dogs, indoor scenes, etc.

false

true

false

117,451

1802.06769

Technique for designing a domain ontology

The article describes the technique for designing a domain ontology, shows the flowchart of algorithm design and example of constructing a fragment of the ontology of the subject area of Computer Science is considered.

false

true

false

90,741

2405.10064

Meta results on data-driven control of nonlinear systems

This note aims to provide a systematic understanding of direct data-driven control, enriching the existing literature not by adding another isolated result, but rather by offering a comprehensive, versatile, and unifying framework that sets the stage for future explorations and applications in this domain. To this end, we formulate the nonlinear design problem from a high-level perspective as a set of desired controlled systems and propose systematic procedures to synthesize data-driven control algorithms that meet the design requirements specified in the desired set. Various examples are presented to demonstrate the comprehensiveness and adaptability of the proposed approach.

false

true

false

454,636

2207.05855

A Conceptual Framework for Using Machine Learning to Support Child Welfare Decisions

Human services systems make key decisions that impact individuals in the society. The U.S. child welfare system makes such decisions, from screening-in hotline reports of suspected abuse or neglect for child protective investigations, placing children in foster care, to returning children to permanent home settings. These complex and impactful decisions on children's lives rely on the judgment of child welfare decisionmakers. Child welfare agencies have been exploring ways to support these decisions with empirical, data-informed methods that include machine learning (ML). This paper describes a conceptual framework for ML to support child welfare decisions. The ML framework guides how child welfare agencies might conceptualize a target problem that ML can solve; vet available administrative data for building ML; formulate and develop ML specifications that mirror relevant populations and interventions the agencies are undertaking; deploy, evaluate, and monitor ML as child welfare context, policy, and practice change over time. Ethical considerations, stakeholder engagement, and avoidance of common pitfalls underpin the framework's impact and success. From abstract to concrete, we describe one application of this framework to support a child welfare decision. This ML framework, though child welfare-focused, is generalizable to solving other public policy problems.

false

true

false

true

false

307,693

1711.00436

Hierarchical Representations for Efficient Architecture Search

We explore efficient neural architecture search methods and show that a simple yet powerful evolutionary algorithm can discover new architectures with excellent performance. Our approach combines a novel hierarchical genetic representation scheme that imitates the modularized design pattern commonly adopted by human experts, and an expressive search space that supports complex topologies. Our algorithm efficiently discovers architectures that outperform a large number of manually designed models for image classification, obtaining top-1 error of 3.6% on CIFAR-10 and 20.3% when transferred to ImageNet, which is competitive with the best existing neural architecture search approaches. We also present results using random search, achieving 0.3% less top-1 accuracy on CIFAR-10 and 0.1% less on ImageNet whilst reducing the search time from 36 hours down to 1 hour.

false

true

false

true

false

true

false

83,716

2302.13577

DuEqNet: Dual-Equivariance Network in Outdoor 3D Object Detection for Autonomous Driving

Outdoor 3D object detection has played an essential role in the environment perception of autonomous driving. In complicated traffic situations, precise object recognition provides indispensable information for prediction and planning in the dynamic system, improving self-driving safety and reliability. However, with the vehicle's veering, the constant rotation of the surrounding scenario makes a challenge for the perception systems. Yet most existing methods have not focused on alleviating the detection accuracy impairment brought by the vehicle's rotation, especially in outdoor 3D detection. In this paper, we propose DuEqNet, which first introduces the concept of equivariance into 3D object detection network by leveraging a hierarchical embedded framework. The dual-equivariance of our model can extract the equivariant features at both local and global levels, respectively. For the local feature, we utilize the graph-based strategy to guarantee the equivariance of the feature in point cloud pillars. In terms of the global feature, the group equivariant convolution layers are adopted to aggregate the local feature to achieve the global equivariance. In the experiment part, we evaluate our approach with different baselines in 3D object detection tasks and obtain State-Of-The-Art performance. According to the results, our model presents higher accuracy on orientation and better prediction efficiency. Moreover, our dual-equivariance strategy exhibits the satisfied plug-and-play ability on various popular object detection frameworks to improve their performance.

false

true

false

348,000

1606.03860

Robust Probabilistic Modeling with Bayesian Data Reweighting

Probabilistic models analyze data by relying on a set of assumptions. Data that exhibit deviations from these assumptions can undermine inference and prediction quality. Robust models offer protection against mismatch between a model's assumptions and reality. We propose a way to systematically detect and mitigate mismatch of a large class of probabilistic models. The idea is to raise the likelihood of each observation to a weight and then to infer both the latent variables and the weights from data. Inferring the weights allows a model to identify observations that match its assumptions and down-weight others. This enables robust inference and improves predictive accuracy. We study four different forms of mismatch with reality, ranging from missing latent groups to structure misspecification. A Poisson factorization analysis of the Movielens 1M dataset shows the benefits of this approach in a practical scenario.

false

true

false

true

false

57,159

2003.08741

A Convolutional Neural Network-based Patent Image Retrieval Method for Design Ideation

The patent database is often used in searches of inspirational stimuli for innovative design opportunities because of its large size, extensive variety and rich design information in patent documents. However, most patent mining research only focuses on textual information and ignores visual information. Herein, we propose a convolutional neural network (CNN)-based patent image retrieval method. The core of this approach is a novel neural network architecture named Dual-VGG that is aimed to accomplish two tasks: visual material type prediction and international patent classification (IPC) class label prediction. In turn, the trained neural network provides the deep features in the image embedding vectors that can be utilized for patent image retrieval and visual mapping. The accuracy of both training tasks and patent image embedding space are evaluated to show the performance of our model. This approach is also illustrated in a case study of robot arm design retrieval. Compared to traditional keyword-based searching and Google image searching, the proposed method discovers more useful visual information for engineering design.

false

true

false

true

false

168,830

1102.2986

Sidon Sequences and Doubly Periodic Two-Dimensional Synchronization Patterns

Sidon sequences and their generalizations have found during the years and especially recently various applications in coding theory. One of the most important applications of these sequences is in the connection of synchronization patterns. A few constructions of two-dimensional synchronization patterns are based on these sequences. In this paper we present sufficient conditions that a two-dimensional synchronization pattern can be transformed into a Sidon sequence. We also present a new construction for Sidon sequences over an alphabet of size q(q-1), where q is a power of a prime.

false

true

false

9,198

2202.10075

ICSML: Industrial Control Systems ML Framework for native inference using IEC 61131-3 code

Industrial Control Systems (ICS) have played a catalytic role in enabling the 4th Industrial Revolution. ICS devices like Programmable Logic Controllers (PLCs), automate, monitor, and control critical processes in industrial, energy, and commercial environments. The convergence of traditional Operational Technology (OT) with Information Technology (IT) has opened a new and unique threat landscape. This has inspired defense research that focuses heavily on Machine Learning (ML) based anomaly detection methods that run on external IT hardware, which means an increase in costs and the further expansion of the threat landscape. To remove this requirement, we introduce the ICS machine learning inference framework (ICSML) which enables executing ML model inference natively on the PLC. ICSML is implemented in IEC 61131-3 code and provides several optimizations to bypass the limitations imposed by the domain-specific languages. Therefore, it works on every PLC without the need for vendor support. ICSML provides a complete set of components for creating full ML models similarly to established ML frameworks. We run a series of benchmarks studying memory and performance, and compare our solution to the TFLite inference framework. At the same time, we develop domain-specific model optimizations to improve the efficiency of ICSML. To demonstrate the abilities of ICSML, we evaluate a case study of a real defense for process-aware attacks targeting a desalination plant.

false

true

false

true

false

true

false

281,412

2003.11540

Learning What to Learn for Video Object Segmentation

Video object segmentation (VOS) is a highly challenging problem, since the target object is only defined during inference with a given first-frame reference mask. The problem of how to capture and utilize this limited target information remains a fundamental research question. We address this by introducing an end-to-end trainable VOS architecture that integrates a differentiable few-shot learning module. This internal learner is designed to predict a powerful parametric model of the target by minimizing a segmentation error in the first frame. We further go beyond standard few-shot learning techniques by learning what the few-shot learner should learn. This allows us to achieve a rich internal representation of the target in the current frame, significantly increasing the segmentation accuracy of our approach. We perform extensive experiments on multiple benchmarks. Our approach sets a new state-of-the-art on the large-scale YouTube-VOS 2018 dataset by achieving an overall score of 81.5, corresponding to a 2.6% relative improvement over the previous best result.

false

true

false

169,637

1906.04649

`Project & Excite' Modules for Segmentation of Volumetric Medical Scans

Fully Convolutional Neural Networks (F-CNNs) achieve state-of-the-art performance for image segmentation in medical imaging. Recently, squeeze and excitation (SE) modules and variations thereof have been introduced to recalibrate feature maps channel- and spatial-wise, which can boost performance while only minimally increasing model complexity. So far, the development of SE has focused on 2D images. In this paper, we propose `Project & Excite' (PE) modules that base upon the ideas of SE and extend them to operating on 3D volumetric images. `Project & Excite' does not perform global average pooling, but squeezes feature maps along different slices of a tensor separately to retain more spatial information that is subsequently used in the excitation step. We demonstrate that PE modules can be easily integrated in 3D U-Net, boosting performance by 5% Dice points, while only increasing the model complexity by 2%. We evaluate the PE module on two challenging tasks, whole-brain segmentation of MRI scans and whole-body segmentation of CT scans. Code: https://github.com/ai-med/squeeze_and_excitation

false

true

false

134,785

2404.17967

SCorP: Statistics-Informed Dense Correspondence Prediction Directly from Unsegmented Medical Images

Statistical shape modeling (SSM) is a powerful computational framework for quantifying and analyzing the geometric variability of anatomical structures, facilitating advancements in medical research, diagnostics, and treatment planning. Traditional methods for shape modeling from imaging data demand significant manual and computational resources. Additionally, these methods necessitate repeating the entire modeling pipeline to derive shape descriptors (e.g., surface-based point correspondences) for new data. While deep learning approaches have shown promise in streamlining the construction of SSMs on new data, they still rely on traditional techniques to supervise the training of the deep networks. Moreover, the predominant linearity assumption of traditional approaches restricts their efficacy, a limitation also inherited by deep learning models trained using optimized/established correspondences. Consequently, representing complex anatomies becomes challenging. To address these limitations, we introduce SCorP, a novel framework capable of predicting surface-based correspondences directly from unsegmented images. By leveraging the shape prior learned directly from surface meshes in an unsupervised manner, the proposed model eliminates the need for an optimized shape model for training supervision. The strong shape prior acts as a teacher and regularizes the feature learning of the student network to guide it in learning image-based features that are predictive of surface correspondences. The proposed model streamlines the training and inference phases by removing the supervision for the correspondence prediction task while alleviating the linearity assumption.

false

true

false

450,075

2108.01265

Memorize, Factorize, or be Na\"ive: Learning Optimal Feature Interaction Methods for CTR Prediction

Click-through rate prediction is one of the core tasks in commercial recommender systems. It aims to predict the probability of a user clicking a particular item given user and item features. As feature interactions bring in non-linearity, they are widely adopted to improve the performance of CTR prediction models. Therefore, effectively modelling feature interactions has attracted much attention in both the research and industry field. The current approaches can generally be categorized into three classes: (1) na\"ive methods, which do not model feature interactions and only use original features; (2) memorized methods, which memorize feature interactions by explicitly viewing them as new features and assigning trainable embeddings; (3) factorized methods, which learn latent vectors for original features and implicitly model feature interactions through factorization functions. Studies have shown that modelling feature interactions by one of these methods alone are suboptimal due to the unique characteristics of different feature interactions. To address this issue, we first propose a general framework called OptInter which finds the most suitable modelling method for each feature interaction. Different state-of-the-art deep CTR models can be viewed as instances of OptInter. To realize the functionality of OptInter, we also introduce a learning algorithm that automatically searches for the optimal modelling method. We conduct extensive experiments on four large datasets. Our experiments show that OptInter improves the best performed state-of-the-art baseline deep CTR models by up to 2.21%. Compared to the memorized method, which also outperforms baselines, we reduce up to 91% parameters. In addition, we conduct several ablation studies to investigate the influence of different components of OptInter. Finally, we provide interpretable discussions on the results of OptInter.

false

true

false

248,974

2202.06771

DS4DH at TREC Health Misinformation 2021: Multi-Dimensional Ranking Models with Transfer Learning and Rank Fusion

This paper describes the work of the Data Science for Digital Health (DS4DH) group at the TREC Health Misinformation Track 2021. The TREC Health Misinformation track focused on the development of retrieval methods that provide relevant, correct and credible information for health related searches on the Web. In our methodology, we used a two-step ranking approach that includes i) a standard retrieval phase, based on BM25 model, and ii) a re-ranking phase, with a pipeline of models focused on the usefulness, supportiveness and credibility dimensions of the retrieved documents. To estimate the usefulness, we classified the initial rank list using pre-trained language models based on the transformers architecture fine-tuned on the MS MARCO corpus. To assess the supportiveness, we utilized BERT-based models fine-tuned on scientific and Wikipedia corpora. Finally, to evaluate the credibility of the documents, we employed a random forest model trained on the Microsoft Credibility dataset combined with a list of credible sites. The resulting ranked lists were then combined using the Reciprocal Rank Fusion algorithm to obtain the final list of useful, supporting and credible documents. Our approach achieved competitive results, being top-2 in the compatibility measurement for the automatic runs. Our findings suggest that integrating automatic ranking models created for each information quality dimension with transfer learning can increase the effectiveness of health-related information retrieval.

false

true

false

280,334

2008.03483

Bidirectional Mapping Generative Adversarial Networks for Brain MR to PET Synthesis

Fusing multi-modality medical images, such as MR and PET, can provide various anatomical or functional information about human body. But PET data is always unavailable due to different reasons such as cost, radiation, or other limitations. In this paper, we propose a 3D end-to-end synthesis network, called Bidirectional Mapping Generative Adversarial Networks (BMGAN), where image contexts and latent vector are effectively used and jointly optimized for brain MR-to-PET synthesis. Concretely, a bidirectional mapping mechanism is designed to embed the semantic information of PET images into the high dimensional latent space. And the 3D DenseU-Net generator architecture and the extensive objective functions are further utilized to improve the visual quality of synthetic results. The most appealing part is that the proposed method can synthesize the perceptually realistic PET images while preserving the diverse brain structures of different subjects. Experimental results demonstrate that the performance of the proposed method outperforms other competitive cross-modality synthesis methods in terms of quantitative measures, qualitative displays, and classification evaluation.

false

true

false

190,918

1505.02619

A Graph Model for Opportunistic Network Coding

Recent advancements in graph-based analysis and solutions of instantly decodable network coding (IDNC) trigger the interest to extend them to more complicated opportunistic network coding (ONC) scenarios, with limited increase in complexity. In this paper, we design a simple IDNC-like graph model for a specific subclass of ONC, by introducing a more generalized definition of its vertices and the notion of vertex aggregation in order to represent the storage of non-instantly-decodable packets in ONC. Based on this representation, we determine the set of pairwise vertex adjacency conditions that can populate this graph with edges so as to guarantee decodability or aggregation for the vertices of each clique in this graph. We then develop the algorithmic procedures that can be applied on the designed graph model to optimize any performance metric for this ONC subclass. A case study on reducing the completion time shows that the proposed framework improves on the performance of IDNC and gets very close to the optimal performance.

false

true

false

42,989

1405.2846

Introduction to Dynamic Unary Encoding

Dynamic unary encoding takes unary encoding to the next level. Every n-bit binary string is an encoding of dynamic unary and every n-bit binary string is encodable by dynamic unary. By utilizing both forms of unary code and a single bit of parity information dynamic unary encoding partitions 2^n non-negative integers into n sets of disjoint cycles of n-bit elements. These cycles have been employed as virtual data sets, binary transforms and as a mathematical object. Characterization of both the cycles and of the cycle spectrum is given. Examples of encoding and decoding algorithms are given. Examples of other constructs utilizing the principles of dynamic unary encoding are presented. The cycle as a mathematical object is demonstrated.

false

true

false

true

33,024

1302.4701

A Receiver-Centric OFCDM Approach with Subcarrier Grouping

In this letter, following a cross-layer design concept, we propose a novel subcarrier grouping technique for Orthogonal Frequency and Code Division Multiplexing (OFCDM) multiuser systems. We adopt a two dimensional (2D) spreading, so as to achieve both frequency- and time-domain channel gain. Furthermore, we enable a receiver-centric approach, where the receiver rather than a potential sender controls the admission decision of the communication establishment. We study the robustness of the proposed scheme in terms of the Bit-Error-Rate (BER) and the outage probability. The derived results indicate that the proposed scheme outperforms the classical OFCDM approach.

false

true

false

22,171

2202.11987

Predicting the impact of treatments over time with uncertainty aware neural differential equations

Predicting the impact of treatments from observational data only still represents a majorchallenge despite recent significant advances in time series modeling. Treatment assignments are usually correlated with the predictors of the response, resulting in a lack of data support for counterfactual predictions and therefore in poor quality estimates. Developments in causal inference have lead to methods addressing this confounding by requiring a minimum level of overlap. However,overlap is difficult to assess and usually notsatisfied in practice. In this work, we propose Counterfactual ODE (CF-ODE), a novel method to predict the impact of treatments continuously over time using Neural Ordinary Differential Equations equipped with uncertainty estimates. This allows to specifically assess which treatment outcomes can be reliably predicted. We demonstrate over several longitudinal data sets that CF-ODE provides more accurate predictions and more reliable uncertainty estimates than previously available methods.

false

true

false

282,078

1108.0155

Reasoning in the OWL 2 Full Ontology Language using First-Order Automated Theorem Proving

OWL 2 has been standardized by the World Wide Web Consortium (W3C) as a family of ontology languages for the Semantic Web. The most expressive of these languages is OWL 2 Full, but to date no reasoner has been implemented for this language. Consistency and entailment checking are known to be undecidable for OWL 2 Full. We have translated a large fragment of the OWL 2 Full semantics into first-order logic, and used automated theorem proving systems to do reasoning based on this theory. The results are promising, and indicate that this approach can be applied in practice for effective OWL reasoning, beyond the capabilities of current Semantic Web reasoners. This is an extended version of a paper with the same title that has been published at CADE 2011, LNAI 6803, pp. 446-460. The extended version provides appendices with additional resources that were used in the reported evaluation.

false

true

false

11,524

2106.12499

Generative Self-training for Cross-domain Unsupervised Tagged-to-Cine MRI Synthesis

Self-training based unsupervised domain adaptation (UDA) has shown great potential to address the problem of domain shift, when applying a trained deep learning model in a source domain to unlabeled target domains. However, while the self-training UDA has demonstrated its effectiveness on discriminative tasks, such as classification and segmentation, via the reliable pseudo-label selection based on the softmax discrete histogram, the self-training UDA for generative tasks, such as image synthesis, is not fully investigated. In this work, we propose a novel generative self-training (GST) UDA framework with continuous value prediction and regression objective for cross-domain image synthesis. Specifically, we propose to filter the pseudo-label with an uncertainty mask, and quantify the predictive confidence of generated images with practical variational Bayes learning. The fast test-time adaptation is achieved by a round-based alternative optimization scheme. We validated our framework on the tagged-to-cine magnetic resonance imaging (MRI) synthesis problem, where datasets in the source and target domains were acquired from different scanners or centers. Extensive validations were carried out to verify our framework against popular adversarial training UDA methods. Results show that our GST, with tagged MRI of test subjects in new target domains, improved the synthesis quality by a large margin, compared with the adversarial training UDA methods.

false

true

false

true

false

true

false

true

false

242,748

2305.02440

Cheaply Evaluating Inference Efficiency Metrics for Autoregressive Transformer APIs

Large language models (LLMs) power many state-of-the-art systems in natural language processing. However, these models are extremely computationally expensive, even at inference time, raising the natural question: when is the extra cost of deploying a larger model worth the anticipated boost in capabilities? Better understanding this tradeoff fundamentally could benefit from an inference efficiency metric that is both (i) easily comparable across models from different providers, and (ii) representative of the true cost of running queries in an isolated performance environment. Unfortunately, access to LLMs today is largely restricted to black-box text generation APIs and raw runtimes measured through this interface do not satisfy these desiderata: model providers can apply various software and hardware optimizations orthogonal to the model, and models served on shared infrastructure are susceptible to performance contention. To circumvent these problems, we propose a new metric for comparing inference efficiency across models. This metric puts models on equal footing as though they were served (i) on uniform hardware and software, and (ii) without performance contention. We call this metric the \emph{idealized runtime}, and we propose a methodology to efficiently estimate this metric for autoregressive Transformer models. We also propose cost-aware variants that incorporate the number of accelerators needed to serve the model. Using these metrics, we compare ten state-of-the-art LLMs to provide the first analysis of inference efficiency-capability tradeoffs; we make several observations from this analysis, including the fact that the superior inference runtime performance of certain APIs is often a byproduct of optimizations within the API rather than the underlying model. Our methodology also facilitates the efficient comparison of different software and hardware stacks.

false

true

false

362,030

2010.05780

Capturing Dynamics of Time-Varying Data via Topology

One approach to understanding complex data is to study its shape through the lens of algebraic topology. While the early development of topological data analysis focused primarily on static data, in recent years, theoretical and applied studies have turned to data that varies in time. A time-varying collection of metric spaces as formed, for example, by a moving school of fish or flock of birds, can contain a vast amount of information. There is often a need to simplify or summarize the dynamic behavior. We provide an introduction to topological summaries of time-varying metric spaces including vineyards [19], crocker plots [56], and multiparameter rank functions [37]. We then introduce a new tool to summarize time-varying metric spaces: a crocker stack. Crocker stacks are convenient for visualization, amenable to machine learning, and satisfy a desirable continuity property which we prove. We demonstrate the utility of crocker stacks for a parameter identification task involving an influential model of biological aggregations [58]. Altogether, we aim to bring the broader applied mathematics community up-to-date on topological summaries of time-varying metric spaces.

false

true

false

true

200,258

2303.17951

FP8 versus INT8 for efficient deep learning inference

Recently, the idea of using FP8 as a number format for neural network training has been floating around the deep learning world. Given that most training is currently conducted with entire networks in FP32, or sometimes FP16 with mixed-precision, the step to having some parts of a network run in FP8 with 8-bit weights is an appealing potential speed-up for the generally costly and time-intensive training procedures in deep learning. A natural question arises regarding what this development means for efficient inference on edge devices. In the efficient inference device world, workloads are frequently executed in INT8. Sometimes going even as low as INT4 when efficiency calls for it. In this whitepaper, we compare the performance for both the FP8 and INT formats for efficient on-device inference. We theoretically show the difference between the INT and FP formats for neural networks and present a plethora of post-training quantization and quantization-aware-training results to show how this theory translates to practice. We also provide a hardware analysis showing that the FP formats are somewhere between 50-180% less efficient in terms of compute in dedicated hardware than the INT format. Based on our research and a read of the research field, we conclude that although the proposed FP8 format could be good for training, the results for inference do not warrant a dedicated implementation of FP8 in favor of INT8 for efficient inference. We show that our results are mostly consistent with previous findings but that important comparisons between the formats have thus far been lacking. Finally, we discuss what happens when FP8-trained networks are converted to INT8 and conclude with a brief discussion on the most efficient way for on-device deployment and an extensive suite of INT8 results for many models.

false

true

false

355,410

1808.01739

Concentration bounds for empirical conditional value-at-risk: The unbounded case

In several real-world applications involving decision making under uncertainty, the traditional expected value objective may not be suitable, as it may be necessary to control losses in the case of a rare but extreme event. Conditional Value-at-Risk (CVaR) is a popular risk measure for modeling the aforementioned objective. We consider the problem of estimating CVaR from i.i.d. samples of an unbounded random variable, which is either sub-Gaussian or sub-exponential. We derive a novel one-sided concentration bound for a natural sample-based CVaR estimator in this setting. Our bound relies on a concentration result for a quantile-based estimator for Value-at-Risk (VaR), which may be of independent interest.

false

true

false

104,638

2303.18176

Affective Computing for Human-Robot Interaction Research: Four Critical Lessons for the Hitchhiker

Social Robotics and Human-Robot Interaction (HRI) research relies on different Affective Computing (AC) solutions for sensing, perceiving and understanding human affective behaviour during interactions. This may include utilising off-the-shelf affect perception models that are pre-trained on popular affect recognition benchmarks and directly applied to situated interactions. However, the conditions in situated human-robot interactions differ significantly from the training data and settings of these models. Thus, there is a need to deepen our understanding of how AC solutions can be best leveraged, customised and applied for situated HRI. This paper, while critiquing the existing practices, presents four critical lessons to be noted by the hitchhiker when applying AC for HRI research. These lessons conclude that: (i) The six basic emotions categories are irrelevant in situated interactions, (ii) Affect recognition accuracy (%) improvements are unimportant, (iii) Affect recognition does not generalise across contexts, and (iv) Affect recognition alone is insufficient for adaptation and personalisation. By describing the background and the context for each lesson, and demonstrating how these lessons have been learnt, this paper aims to enable the hitchhiker to successfully and insightfully leverage AC solutions for advancing HRI research.

true

false

true

false

355,489

1906.03609

Distilling Object Detectors with Fine-grained Feature Imitation

State-of-the-art CNN based recognition models are often computationally prohibitive to deploy on low-end devices. A promising high level approach tackling this limitation is knowledge distillation, which let small student model mimic cumbersome teacher model's output to get improved generalization. However, related methods mainly focus on simple task of classification while do not consider complex tasks like object detection. We show applying the vanilla knowledge distillation to detection model gets minor gain. To address the challenge of distilling knowledge in detection model, we propose a fine-grained feature imitation method exploiting the cross-location discrepancy of feature response. Our intuition is that detectors care more about local near object regions. Thus the discrepancy of feature response on the near object anchor locations reveals important information of how teacher model tends to generalize. We design a novel mechanism to estimate those locations and let student model imitate the teacher on them to get enhanced performance. We first validate the idea on a developed lightweight toy detector which carries simplest notion of current state-of-the-art anchor based detection models on challenging KITTI dataset, our method generates up to 15% boost of mAP for the student model compared to the non-imitated counterpart. We then extensively evaluate the method with Faster R-CNN model under various scenarios with common object detection benchmark of Pascal VOC and COCO, imitation alleviates up to 74% performance drop of student model compared to teacher. Codes released at https://github.com/twangnh/Distilling-Object-Detectors

false

true

false

true

false

true

false

134,434

2208.04273

Improving performance in multi-objective decision-making in Bottles environments with soft maximin approaches

Balancing multiple competing and conflicting objectives is an essential task for any artificial intelligence tasked with satisfying human values or preferences. Conflict arises both from misalignment between individuals with competing values, but also between conflicting value systems held by a single human. Starting with principle of loss-aversion, we designed a set of soft maximin function approaches to multi-objective decision-making. Bench-marking these functions in a set of previously-developed environments, we found that one new approach in particular, 'split-function exp-log loss aversion' (SFELLA), learns faster than the state of the art thresholded alignment objective method (Vamplew et al, 2021) on three of four tasks it was tested on, and achieved the same optimal performance after learning. SFELLA also showed relative robustness improvements against changes in objective scale, which may highlight an advantage dealing with distribution shifts in the environment dynamics. Due to publishing rules, further work could not be presented in the preprint, but in the final published version, we will further compare SFELLA to the multi-objective reward exponentials (MORE) approach (Rolf, 2020), demonstrating that SFELLA performs similarly to MORE in a simple previously-described foraging task, but in a modified foraging environment with a new resource that was not depleted as the agent worked, SFELLA collected more of the new resource with very little cost incurred in terms of the old resource. Overall, we found SFELLA useful for avoiding problems that sometimes occur with a thresholded approach, and more reward-responsive than MORE while retaining its conservative, loss-averse incentive structure.

false

true

false

312,052

2207.03960

Detection of Furigana Text in Images

Furigana are pronunciation notes used in Japanese writing. Being able to detect these can help improve optical character recognition (OCR) performance or make more accurate digital copies of Japanese written media by correctly displaying furigana. This project focuses on detecting furigana in Japanese books and comics. While there has been research into the detection of Japanese text in general, there are currently no proposed methods for detecting furigana. We construct a new dataset containing Japanese written media and annotations of furigana. We propose an evaluation metric for such data which is similar to the evaluation protocols used in object detection except that it allows groups of objects to be labeled by one annotation. We propose a method for detection of furigana that is based on mathematical morphology and connected component analysis. We evaluate the detections of the dataset and compare different methods for text extraction. We also evaluate different types of images such as books and comics individually and discuss the challenges of each type of image. The proposed method reaches an F1-score of 76\% on the dataset. The method performs well on regular books, but less so on comics, and books of irregular format. Finally, we show that the proposed method can improve the performance of OCR by 5\% on the manga109 dataset. Source code is available via \texttt{\url{https://github.com/nikolajkb/FuriganaDetection}}

false

true

false

307,032

1912.12712

Haptic communication optimises joint decisions and affords implicit confidence sharing

Group decisions can outperform the choices of the best individual group members. Previous research suggested that optimal group decisions require individuals to communicate explicitly (e.g., verbally) their confidence levels. Our study addresses the untested hypothesis that implicit communication using a sensorimotor channel -- haptic coupling -- may afford optimal group decisions, too. We report that haptically coupled dyads solve a perceptual discrimination task more accurately than their best individual members; and five times faster than dyads using explicit communication. Furthermore, our computational analyses indicate that the haptic channel affords implicit confidence sharing. We found that dyads take leadership over the choice and communicate their confidence in it by modulating both the timing and the force of their movements. Our findings may pave the way to negotiation technologies using fast sensorimotor communication to solve problems in groups.

true

false

true

false

true

false

158,911

2411.03740

Human-in-the-Loop Feature Selection Using Interpretable Kolmogorov-Arnold Network-based Double Deep Q-Network

Feature selection is critical for improving the performance and interpretability of machine learning models, particularly in high-dimensional spaces where complex feature interactions can reduce accuracy and increase computational demands. Existing approaches often rely on static feature subsets or manual intervention, limiting adaptability and scalability. However, dynamic, per-instance feature selection methods and model-specific interpretability in reinforcement learning remain underexplored. This study proposes a human-in-the-loop (HITL) feature selection framework integrated into a Double Deep Q-Network (DDQN) using a Kolmogorov-Arnold Network (KAN). Our novel approach leverages simulated human feedback and stochastic distribution-based sampling, specifically Beta, to iteratively refine feature subsets per data instance, improving flexibility in feature selection. The KAN-DDQN achieved notable test accuracies of 93% on MNIST and 83% on FashionMNIST, outperforming conventional MLP-DDQN models by up to 9%. The KAN-based model provided high interpretability via symbolic representation while using 4 times fewer neurons in the hidden layer than MLPs did. Comparatively, the models without feature selection achieved test accuracies of only 58% on MNIST and 64% on FashionMNIST, highlighting significant gains with our framework. Pruning and visualization further enhanced model transparency by elucidating decision pathways. These findings present a scalable, interpretable solution for feature selection that is suitable for applications requiring real-time, adaptive decision-making with minimal human oversight.

true

false

true

false

506,019

1909.07877

Multi-mapping Image-to-Image Translation via Learning Disentanglement

Recent advances of image-to-image translation focus on learning the one-to-many mapping from two aspects: multi-modal translation and multi-domain translation. However, the existing methods only consider one of the two perspectives, which makes them unable to solve each other's problem. To address this issue, we propose a novel unified model, which bridges these two objectives. First, we disentangle the input images into the latent representations by an encoder-decoder architecture with a conditional adversarial training in the feature space. Then, we encourage the generator to learn multi-mappings by a random cross-domain translation. As a result, we can manipulate different parts of the latent representations to perform multi-modal and multi-domain translations simultaneously. Experiments demonstrate that our method outperforms state-of-the-art methods.

false

true

false

145,804

2104.09382

A Joint Energy and Latency Framework for Transfer Learning over 5G Industrial Edge Networks

In this paper, we propose a transfer learning (TL)-enabled edge-CNN framework for 5G industrial edge networks with privacy-preserving characteristic. In particular, the edge server can use the existing image dataset to train the CNN in advance, which is further fine-tuned based on the limited datasets uploaded from the devices. With the aid of TL, the devices that are not participating in the training only need to fine-tune the trained edge-CNN model without training from scratch. Due to the energy budget of the devices and the limited communication bandwidth, a joint energy and latency problem is formulated, which is solved by decomposing the original problem into an uploading decision subproblem and a wireless bandwidth allocation subproblem. Experiments using ImageNet demonstrate that the proposed TL-enabled edge-CNN framework can achieve almost 85% prediction accuracy of the baseline by uploading only about 1% model parameters, for a compression ratio of 32 of the autoencoder.

false

true

false

true

231,219

2412.02109

Direct Coloring for Self-Supervised Enhanced Feature Decoupling

The success of self-supervised learning (SSL) has been the focus of multiple recent theoretical and empirical studies, including the role of data augmentation (in feature decoupling) as well as complete and dimensional representation collapse. While complete collapse is well-studied and addressed, dimensional collapse has only gain attention and addressed in recent years mostly using variants of redundancy reduction (aka whitening) techniques. In this paper, we further explore a complementary approach to whitening via feature decoupling for improved representation learning while avoiding representation collapse. In particular, we perform feature decoupling by early promotion of useful features via careful feature coloring. The coloring technique is developed based on a Bayesian prior of the augmented data, which is inherently encoded for feature decoupling. We show that our proposed framework is complementary to the state-of-the-art techniques, while outperforming both contrastive and recent non-contrastive methods. We also study the different effects of coloring approach to formulate it as a general complementary technique along with other baselines.

false

true

false

513,384

1912.05345

Severity Detection Tool for Patients with Infectious Disease

Hand, foot and mouth disease (HFMD) and tetanus are serious infectious diseases in low and middle income countries. Tetanus in particular has a high mortality rate and its treatment is resource-demanding. Furthermore, HFMD often affects a large number of infants and young children. As a result, its treatment consumes enormous healthcare resources, especially when outbreaks occur. Autonomic nervous system dysfunction (ANSD) is the main cause of death for both HFMD and tetanus patients. However, early detection of ANSD is a difficult and challenging problem. In this paper, we aim to provide a proof-of-principle to detect the ANSD level automatically by applying machine learning techniques to physiological patient data, such as electrocardiogram (ECG) and photoplethysmogram (PPG) waveforms, which can be collected using low-cost wearable sensors. Efficient features are extracted that encode variations in the waveforms in the time and frequency domains. A support vector machine is employed to classify the ANSD levels. The proposed approach is validated on multiple datasets of HFMD and tetanus patients in Vietnam. Results show that encouraging performance is achieved in classifying ANSD levels. Moreover, the proposed features are simple, more generalisable and outperformed the standard heart rate variability (HRV) analysis. The proposed approach would facilitate both the diagnosis and treatment of infectious diseases in low and middle income countries, and thereby improve overall patient care.

false

true

false

true

false

157,086

1804.07209

NAIS-Net: Stable Deep Networks from Non-Autonomous Differential Equations

This paper introduces Non-Autonomous Input-Output Stable Network(NAIS-Net), a very deep architecture where each stacked processing block is derived from a time-invariant non-autonomous dynamical system. Non-autonomy is implemented by skip connections from the block input to each of the unrolled processing stages and allows stability to be enforced so that blocks can be unrolled adaptively to a pattern-dependent processing depth. NAIS-Net induces non-trivial, Lipschitz input-output maps, even for an infinite unroll length. We prove that the network is globally asymptotically stable so that for every initial condition there is exactly one input-dependent equilibrium assuming $tanh$ units, and incrementally stable for ReL units. An efficient implementation that enforces the stability under derived conditions for both fully-connected and convolutional layers is also presented. Experimental results show how NAIS-Net exhibits stability in practice, yielding a significant reduction in generalization gap compared to ResNets.

false

true

false

true

false

95,478

1702.05528

Degrees of Freedom in Cached MIMO Relay Networks With Multiple Base Stations

The ability of physical layer relay caching to increase the degrees of freedom (DoF) of a single cell was recently illustrated. In this paper, we extend this result to the case of multiple cells in which a caching relay is shared among multiple non-cooperative base stations (BSs). In particular, we show that a large DoF gain can be achieved by exploiting the benefits of having a shared relay that cooperates with the BSs. We first propose a cache-assisted relaying protocol that improves the cooperation opportunity between the BSs and the relay. Next, we consider the cache content placement problem that aims to design the cache content at the relay such that the DoF gain is maximized. We propose an optimal algorithm and a near-optimal low-complexity algorithm for the cache content placement problem. Simulation results show significant improvement in the DoF gain using the proposed relay-caching protocol.

false

true

false

68,415

2202.08132

Prospect Pruning: Finding Trainable Weights at Initialization using Meta-Gradients

Pruning neural networks at initialization would enable us to find sparse models that retain the accuracy of the original network while consuming fewer computational resources for training and inference. However, current methods are insufficient to enable this optimization and lead to a large degradation in model performance. In this paper, we identify a fundamental limitation in the formulation of current methods, namely that their saliency criteria look at a single step at the start of training without taking into account the trainability of the network. While pruning iteratively and gradually has been shown to improve pruning performance, explicit consideration of the training stage that will immediately follow pruning has so far been absent from the computation of the saliency criterion. To overcome the short-sightedness of existing methods, we propose Prospect Pruning (ProsPr), which uses meta-gradients through the first few steps of optimization to determine which weights to prune. ProsPr combines an estimate of the higher-order effects of pruning on the loss and the optimization trajectory to identify the trainable sub-network. Our method achieves state-of-the-art pruning performance on a variety of vision classification tasks, with less data and in a single shot compared to existing pruning-at-initialization methods.

false

true

false

280,773

2111.03476

A Variational U-Net for Weather Forecasting

Not only can discovering patterns and insights from atmospheric data enable more accurate weather predictions, but it may also provide valuable information to help tackle climate change. Weather4cast is an open competition that aims to evaluate machine learning algorithms' capability to predict future atmospheric states. Here, we describe our third-place solution to Weather4cast. We present a novel Variational U-Net that combines a Variational Autoencoder's ability to consider the probabilistic nature of data with a U-Net's ability to recover fine-grained details. This solution is an evolution from our fourth-place solution to Traffic4cast 2020 with many commonalities, suggesting its applicability to vastly different domains, such as weather and traffic.

false

true

false

265,185

1607.06996

Scaling Up Sparse Support Vector Machines by Simultaneous Feature and Sample Reduction

Sparse support vector machine (SVM) is a popular classification technique that can simultaneously learn a small set of the most interpretable features and identify the support vectors. It has achieved great successes in many real-world applications. However, for large-scale problems involving a huge number of samples and ultra-high dimensional features, solving sparse SVMs remains challenging. By noting that sparse SVMs induce sparsities in both feature and sample spaces, we propose a novel approach, which is based on accurate estimations of the primal and dual optima of sparse SVMs, to simultaneously identify the inactive features and samples that are guaranteed to be irrelevant to the outputs. Thus, we can remove the identified inactive samples and features from the training phase, leading to substantial savings in the computational cost without sacrificing the accuracy. Moreover, we show that our method can be extended to multi-class sparse support vector machines. To the best of our knowledge, the proposed method is the \emph{first} \emph{static} feature and sample reduction method for sparse SVMs and multi-class sparse SVMs. Experiments on both synthetic and real data sets demonstrate that our approach significantly outperforms state-of-the-art methods and the speedup gained by our approach can be orders of magnitude.

false

true

false

58,960

1406.4567

Two Boolean functions with five-valued Walsh spectra and high nonlinearity

For cryptographic systems the method of confusion and diffusion is used as a fundamental technique to achieve security. Confusion is reflected in nonlinearity of certain Boolean functions describing the cryptographic transformation. In this paper, we present two balanced boolean functions which have low Walsh spectra and high nonlinearity. In the proof of the nonlinearity, a new method for evaluating some exponential sums over finite fields was provided.

false

true

false

33,948

1610.03612

The Analysis of Local Motion and Deformation in Image Sequences Inspired by Physical Electromagnetic Interaction

In order to analyze the moving and deforming of the objects in image sequence, a novel way is presented to analyze the local changes of object edges between two related images (such as two adjacent frames in a video sequence), which is inspired by the physical electromagnetic interaction. The changes of edge between adjacent frames in sequences are analyzed by simulation of virtual current interaction, which can reflect the change of the object's position or shape. The virtual current along the main edge line is proposed based on the significant edge extraction. Then the virtual interaction between the current elements in the two related images is studied by imitating the interaction between physical current-carrying wires. The experimental results prove that the distribution of magnetic forces on the current elements in one image applied by the other can reflect the local change of edge lines from one image to the other, which is important in further analysis.

false

true

false

62,270

2107.03356

M-FAC: Efficient Matrix-Free Approximations of Second-Order Information

Efficiently approximating local curvature information of the loss function is a key tool for optimization and compression of deep neural networks. Yet, most existing methods to approximate second-order information have high computational or storage costs, which can limit their practicality. In this work, we investigate matrix-free, linear-time approaches for estimating Inverse-Hessian Vector Products (IHVPs) for the case when the Hessian can be approximated as a sum of rank-one matrices, as in the classic approximation of the Hessian by the empirical Fisher matrix. We propose two new algorithms as part of a framework called M-FAC: the first algorithm is tailored towards network compression and can compute the IHVP for dimension $d$, if the Hessian is given as a sum of $m$ rank-one matrices, using $O(dm^2)$ precomputation, $O(dm)$ cost for computing the IHVP, and query cost $O(m)$ for any single element of the inverse Hessian. The second algorithm targets an optimization setting, where we wish to compute the product between the inverse Hessian, estimated over a sliding window of optimization steps, and a given gradient direction, as required for preconditioned SGD. We give an algorithm with cost $O(dm + m^2)$ for computing the IHVP and $O(dm + m^3)$ for adding or removing any gradient from the sliding window. These two algorithms yield state-of-the-art results for network pruning and optimization with lower computational overhead relative to existing second-order methods. Implementations are available at [9] and [17].

false

true

false

245,137

2412.08114

Modeling Latent Non-Linear Dynamical System over Time Series

We study the problem of modeling a non-linear dynamical system when given a time series by deriving equations directly from the data. Despite the fact that time series data are given as input, models for dynamics and estimation algorithms that incorporate long-term temporal dependencies are largely absent from existing studies. In this paper, we introduce a latent state to allow time-dependent modeling and formulate this problem as a dynamics estimation problem in latent states. We face multiple technical challenges, including (1) modeling latent non-linear dynamics and (2) solving circular dependencies caused by the presence of latent states. To tackle these challenging problems, we propose a new method, Latent Non-Linear equation modeling (LaNoLem), that can model a latent non-linear dynamical system and a novel alternating minimization algorithm for effectively estimating latent states and model parameters. In addition, we introduce criteria to control model complexity without human intervention. Compared with the state-of-the-art model, LaNoLem achieves competitive performance for estimating dynamics while outperforming other methods in prediction.

false

true

false

515,948

1511.02942

ExtraPush for convex smooth decentralized optimization over directed networks

In this note, we extend the algorithms Extra and subgradient-push to a new algorithm ExtraPush for consensus optimization with convex differentiable objective functions over a directed network. When the stationary distribution of the network can be computed in advance}, we propose a simplified algorithm called Normalized ExtraPush. Just like Extra, both ExtraPush and Normalized ExtraPush can iterate with a fixed step size. But unlike Extra, they can take a column-stochastic mixing matrix, which is not necessarily doubly stochastic. Therefore, they remove the undirected-network restriction of Extra. Subgradient-push, while also works for directed networks, is slower on the same type of problem because it must use a sequence of diminishing step sizes. We present preliminary analysis for ExtraPush under a bounded sequence assumption. For Normalized ExtraPush, we show that it naturally produces a bounded, linearly convergent sequence provided that the objective function is strongly convex. In our numerical experiments, ExtraPush and Normalized ExtraPush performed similarly well. They are significantly faster than subgradient-push, even when we hand-optimize the step sizes for the latter.

false

true

false

true

48,699

1301.7408

Context-Specific Approximation in Probabilistic Inference

There is evidence that the numbers in probabilistic inference don't really matter. This paper considers the idea that we can make a probabilistic model simpler by making fewer distinctions. Unfortunately, the level of a Bayesian network seems too coarse; it is unlikely that a parent will make little difference for all values of the other parents. In this paper we consider an approximation scheme where distinctions can be ignored in some contexts, but not in other contexts. We elaborate on a notion of a parent context that allows a structured context-specific decomposition of a probability distribution and the associated probabilistic inference scheme called probabilistic partial evaluation (Poole 1997). This paper shows a way to simplify a probabilistic model by ignoring distinctions which have similar probabilities, a method to exploit the simpler model, a bound on the resulting errors, and some preliminary empirical results on simple networks.

false

true

false

21,641

2411.10048

Physics-informed neural networks need a physicist to be accurate: the case of mass and heat transport in Fischer-Tropsch catalyst particles

Physics-Informed Neural Networks (PINNs) have emerged as an influential technology, merging the swift and automated capabilities of machine learning with the precision and dependability of simulations grounded in theoretical physics. PINNs are often employed to solve algebraic or differential equations to replace some or even all steps of multi-stage computational workflows, leading to their significant speed-up. However, wide adoption of PINNs is still hindered by reliability issues, particularly at extreme ends of the input parameter ranges. In this study, we demonstrate this in the context of a system of coupled non-linear differential reaction-diffusion and heat transfer equations related to Fischer-Tropsch synthesis, which are solved by a finite-difference method with a PINN used in evaluating their source terms. It is shown that the testing strategies traditionally used to assess the accuracy of neural networks as function approximators can overlook the peculiarities which ultimately cause instabilities of the finite-difference solver. We propose a domain knowledge-based modifications to the PINN architecture ensuring its correct asymptotic behavior. When combined with an improved numerical scheme employed as an initial guess generator, the proposed modifications are shown to recover the overall stability of the simulations, while preserving the speed-up brought by PINN as the workflow component. We discuss the possible applications of the proposed hybrid transport equation solver in context of chemical reactors simulations.

false

true

false

true

false

508,474

2311.04066

Can CLIP Help Sound Source Localization?

Large-scale pre-trained image-text models demonstrate remarkable versatility across diverse tasks, benefiting from their robust representational capabilities and effective multimodal alignment. We extend the application of these models, specifically CLIP, to the domain of sound source localization. Unlike conventional approaches, we employ the pre-trained CLIP model without explicit text input, relying solely on the audio-visual correspondence. To this end, we introduce a framework that translates audio signals into tokens compatible with CLIP's text encoder, yielding audio-driven embeddings. By directly using these embeddings, our method generates audio-grounded masks for the provided audio, extracts audio-grounded image features from the highlighted regions, and aligns them with the audio-driven embeddings using the audio-visual correspondence objective. Our findings suggest that utilizing pre-trained image-text models enable our model to generate more complete and compact localization maps for the sounding objects. Extensive experiments show that our method outperforms state-of-the-art approaches by a significant margin.

false

true

false

true

false

true

false

true

406,089

1410.4155

Access Policy Design for Cognitive Secondary Users under a Primary Type-I HARQ Process

In this paper, an underlay cognitive radio network that consists of an arbitrary number of secondary users (SU) is considered, in which the primary user (PU) employs Type-I Hybrid Automatic Repeat Request (HARQ). Exploiting the redundancy in PU retransmissions, each SU receiver applies forward interference cancelation to remove a successfully decoded PU message in the subsequent PU retransmissions. The knowledge of the PU message state at the SU receivers and the ACK/NACK message from the PU receiver are sent back to the transmitters. With this approach and using a Constrained Markov Decision Process (CMDP) model and Constrained Multi-agent MDP (CMMDP), centralized and decentralized optimum access policies for SUs are proposed to maximize their average sum throughput under a PU throughput constraint. In the decentralized case, the channel access decision of each SU is unknown to the other SU. Numerical results demonstrate the benefits of the proposed policies in terms of sum throughput of SUs. The results also reveal that the centralized access policy design outperforms the decentralized design especially when the PU can tolerate a low average long term throughput. Finally, the difficulties in decentralized access policy design with partial state information are discussed.

false

true

false

36,771

0810.1119

Gaussian Belief Propagation for Solving Systems of Linear Equations: Theory and Application

The canonical problem of solving a system of linear equations arises in numerous contexts in information theory, communication theory, and related fields. In this contribution, we develop a solution based upon Gaussian belief propagation (GaBP) that does not involve direct matrix inversion. The iterative nature of our approach allows for a distributed message-passing implementation of the solution algorithm. We address the properties of the GaBP solver, including convergence, exactness, computational complexity, message-passing efficiency and its relation to classical solution methods. We use numerical examples and applications, like linear detection, to illustrate these properties through the use of computer simulations. This empirical study demonstrates the attractiveness (e.g., faster convergence rate) of the proposed GaBP solver in comparison to conventional linear-algebraic iterative solution methods.

false

true

false

2,458

2210.17190

IITD at the WANLP 2022 Shared Task: Multilingual Multi-Granularity Network for Propaganda Detection

We present our system for the two subtasks of the shared task on propaganda detection in Arabic, part of WANLP'2022. Subtask 1 is a multi-label classification problem to find the propaganda techniques used in a given tweet. Our system for this task uses XLM-R to predict probabilities for the target tweet to use each of the techniques. In addition to finding the techniques, Subtask 2 further asks to identify the textual span for each instance of each technique that is present in the tweet; the task can be modeled as a sequence tagging problem. We use a multi-granularity network with mBERT encoder for Subtask 2. Overall, our system ranks second for both subtasks (out of 14 and 3 participants, respectively). Our empirical analysis show that it does not help to use a much larger English corpus annotated with propaganda techniques, regardless of whether used in English or after translation to Arabic.

false

true

false

true

false

327,602

2107.03227

Scalable Data Balancing for Unlabeled Satellite Imagery

Data imbalance is a ubiquitous problem in machine learning. In large scale collected and annotated datasets, data imbalance is either mitigated manually by undersampling frequent classes and oversampling rare classes, or planned for with imputation and augmentation techniques. In both cases balancing data requires labels. In other words, only annotated data can be balanced. Collecting fully annotated datasets is challenging, especially for large scale satellite systems such as the unlabeled NASA's 35 PB Earth Imagery dataset. Although the NASA Earth Imagery dataset is unlabeled, there are implicit properties of the data source that we can rely on to hypothesize about its imbalance, such as distribution of land and water in the case of the Earth's imagery. We present a new iterative method to balance unlabeled data. Our method utilizes image embeddings as a proxy for image labels that can be used to balance data, and ultimately when trained increases overall accuracy.

false

true

false

true

false

true

false

245,097

2305.01997

Extraction of volumetric indices from echocardiography: which deep learning solution for clinical use?

Deep learning-based methods have spearheaded the automatic analysis of echocardiographic images, taking advantage of the publication of multiple open access datasets annotated by experts (CAMUS being one of the largest public databases). However, these models are still considered unreliable by clinicians due to unresolved issues concerning i) the temporal consistency of their predictions, and ii) their ability to generalize across datasets. In this context, we propose a comprehensive comparison between the current best performing methods in medical/echocardiographic image segmentation, with a particular focus on temporal consistency and cross-dataset aspects. We introduce a new private dataset, named CARDINAL, of apical two-chamber and apical four-chamber sequences, with reference segmentation over the full cardiac cycle. We show that the proposed 3D nnU-Net outperforms alternative 2D and recurrent segmentation methods. We also report that the best models trained on CARDINAL, when tested on CAMUS without any fine-tuning, still manage to perform competitively with respect to prior methods. Overall, the experimental results suggest that with sufficient training data, 3D nnU-Net could become the first automated tool to finally meet the standards of an everyday clinical device.

false

true

false

true

false

true

false

361,872