April 2, 2020

3065 words 15 mins read

Paper Group ANR 344

SS-Auto: A Single-Shot, Automatic Structured Weight Pruning Framework of DNNs with Ultra-High Efficiency. Processing topical queries on images of historical newspaper pages. Learning Functionally Decomposed Hierarchies for Continuous Control Tasks. Hierarchical Reinforcement Learning as a Model of Human Task Interleaving. Effect of Confidence and E …

SS-Auto: A Single-Shot, Automatic Structured Weight Pruning Framework of DNNs with Ultra-High Efficiency


Title	SS-Auto: A Single-Shot, Automatic Structured Weight Pruning Framework of DNNs with Ultra-High Efficiency
Authors	Zhengang Li, Yifan Gong, Xiaolong Ma, Sijia Liu, Mengshu Sun, Zheng Zhan, Zhenglun Kong, Geng Yuan, Yanzhi Wang
Abstract	Structured weight pruning is a representative model compression technique of DNNs for hardware efficiency and inference accelerations. Previous works in this area leave great space for improvement since sparse structures with combinations of different structured pruning schemes are not exploited fully and efficiently. To mitigate the limitations, we propose SS-Auto, a single-shot, automatic structured pruning framework that can achieve row pruning and column pruning simultaneously. We adopt soft constraint-based formulation to alleviate the strong non-convexity of l0-norm constraints used in state-of-the-art ADMM-based methods for faster convergence and fewer hyperparameters. Instead of solving the problem directly, a Primal-Proximal solution is proposed to avoid the pitfall of penalizing all weights equally, thereby enhancing the accuracy. Extensive experiments on CIFAR-10 and CIFAR-100 datasets demonstrate that the proposed framework can achieve ultra-high pruning rates while maintaining accuracy. Furthermore, significant inference speedup has been observed from the proposed framework through actual measurements on the smartphone.
Tasks	Model Compression
Published	2020-01-23
URL	https://arxiv.org/abs/2001.08839v1
PDF	https://arxiv.org/pdf/2001.08839v1.pdf
PWC	https://paperswithcode.com/paper/ss-auto-a-single-shot-automatic-structured
Repo
Framework

Processing topical queries on images of historical newspaper pages


Title	Processing topical queries on images of historical newspaper pages
Authors	José E. B. Maia, Gildácio J. de A. Sá
Abstract	Historical newspapers are a source of research for the human and social sciences. However, these image collections are difficult to read by machine due to the low quality of the print, the lack of standardization of the pages in addition to the low quality photograph of some files. This paper presents the processing model of a topic navigation system in historical newspaper page images. The general procedure consists of four modules which are: segmentation of text sub-images and text extraction, preprocessing and representation, induced topic extraction and representation, and document viewing and retrieval interface. The algorithmic and technological approaches of each module are described and the initial test results about a collection covering a range of 28 years are presented.
Tasks
Published	2020-02-20
URL	https://arxiv.org/abs/2002.08500v1
PDF	https://arxiv.org/pdf/2002.08500v1.pdf
PWC	https://paperswithcode.com/paper/processing-topical-queries-on-images-of
Repo
Framework

Learning Functionally Decomposed Hierarchies for Continuous Control Tasks


Title	Learning Functionally Decomposed Hierarchies for Continuous Control Tasks
Authors	Lukas Jendele, Sammy Christen, Emre Aksan, Otmar Hilliges
Abstract	Solving long-horizon sequential decision making tasks in environments with sparse rewards is a longstanding problem in reinforcement learning (RL) research. Hierarchical Reinforcement Learning (HRL) has held the promise to enhance the capabilities of RL agents via operation on different levels of temporal abstraction. Despite the success of recent works in dealing with inherent nonstationarity and sample complexity, it remains difficult to generalize to unseen environments and to transfer different layers of the policy to other agents. In this paper, we propose a novel HRL architecture, Hierarchical Decompositional Reinforcement Learning (HiDe), which allows decomposition of the hierarchical layers into independent subtasks, yet allows for joint training of all layers in end-to-end manner. The main insight is to combine a control policy on a lower level with an image-based planning policy on a higher level. We evaluate our method on various complex continuous control tasks, demonstrating that generalization across environments and transfer of higher level policies, such as from a simple ball to a complex humanoid, can be achieved. See videos https://sites.google.com/view/hide-rl.
Tasks	Continuous Control, Decision Making, Hierarchical Reinforcement Learning
Published	2020-02-14
URL	https://arxiv.org/abs/2002.05954v1
PDF	https://arxiv.org/pdf/2002.05954v1.pdf
PWC	https://paperswithcode.com/paper/learning-functionally-decomposed-hierarchies-1
Repo
Framework

Hierarchical Reinforcement Learning as a Model of Human Task Interleaving


Title	Hierarchical Reinforcement Learning as a Model of Human Task Interleaving
Authors	Christoph Gebhardt, Antti Oulasvirta, Otmar Hilliges
Abstract	How do people decide how long to continue in a task, when to switch, and to which other task? Understanding the mechanisms that underpin task interleaving is a long-standing goal in the cognitive sciences. Prior work suggests greedy heuristics and a policy maximizing the marginal rate of return. However, it is unclear how such a strategy would allow for adaptation to everyday environments that offer multiple tasks with complex switch costs and delayed rewards. Here we develop a hierarchical model of supervisory control driven by reinforcement learning (RL). The supervisory level learns to switch using task-specific approximate utility estimates, which are computed on the lower level. A hierarchically optimal value function decomposition can be learned from experience, even in conditions with multiple tasks and arbitrary and uncertain reward and cost structures. The model reproduces known empirical effects of task interleaving. It yields better predictions of individual-level data than a myopic baseline in a six-task problem (N=211). The results support hierarchical RL as a plausible model of task interleaving.
Tasks	Hierarchical Reinforcement Learning
Published	2020-01-04
URL	https://arxiv.org/abs/2001.02122v1
PDF	https://arxiv.org/pdf/2001.02122v1.pdf
PWC	https://paperswithcode.com/paper/hierarchical-reinforcement-learning-as-a
Repo
Framework

Effect of Confidence and Explanation on Accuracy and Trust Calibration in AI-Assisted Decision Making


Title	Effect of Confidence and Explanation on Accuracy and Trust Calibration in AI-Assisted Decision Making
Authors	Yunfeng Zhang, Q. Vera Liao, Rachel K. E. Bellamy
Abstract	Today, AI is being increasingly used to help human experts make decisions in high-stakes scenarios. In these scenarios, full automation is often undesirable, not only due to the significance of the outcome, but also because human experts can draw on their domain knowledge complementary to the model’s to ensure task success. We refer to these scenarios as AI-assisted decision making, where the individual strengths of the human and the AI come together to optimize the joint decision outcome. A key to their success is to appropriately \textit{calibrate} human trust in the AI on a case-by-case basis; knowing when to trust or distrust the AI allows the human expert to appropriately apply their knowledge, improving decision outcomes in cases where the model is likely to perform poorly. This research conducts a case study of AI-assisted decision making in which humans and AI have comparable performance alone, and explores whether features that reveal case-specific model information can calibrate trust and improve the joint performance of the human and AI. Specifically, we study the effect of showing confidence score and local explanation for a particular prediction. Through two human experiments, we show that confidence score can help calibrate people’s trust in an AI model, but trust calibration alone is not sufficient to improve AI-assisted decision making, which may also depend on whether the human can bring in enough unique knowledge to complement the AI’s errors. We also highlight the problems in using local explanation for AI-assisted decision making scenarios and invite the research community to explore new approaches to explainability for calibrating human trust in AI.
Tasks	Calibration, Decision Making
Published	2020-01-07
URL	https://arxiv.org/abs/2001.02114v1
PDF	https://arxiv.org/pdf/2001.02114v1.pdf
PWC	https://paperswithcode.com/paper/effect-of-confidence-and-explanation-on
Repo
Framework

Simple Interactive Image Segmentation using Label Propagation through kNN graphs


Title	Simple Interactive Image Segmentation using Label Propagation through kNN graphs
Authors	Fabricio Aparecido Breve
Abstract	Many interactive image segmentation techniques are based on semi-supervised learning. The user may label some pixels from each object and the SSL algorithm will propagate the labels from the labeled to the unlabeled pixels, finding object boundaries. This paper proposes a new SSL graph-based interactive image segmentation approach, using undirected and unweighted kNN graphs, from which the unlabeled nodes receive contributions from other nodes (either labeled or unlabeled). It is simpler than many other techniques, but it still achieves significant classification accuracy in the image segmentation task. Computer simulations are performed using some real-world images, extracted from the Microsoft GrabCut dataset. The segmentation results show the effectiveness of the proposed approach.
Tasks	Semantic Segmentation
Published	2020-02-13
URL	https://arxiv.org/abs/2002.05708v1
PDF	https://arxiv.org/pdf/2002.05708v1.pdf
PWC	https://paperswithcode.com/paper/simple-interactive-image-segmentation-using
Repo
Framework

Gimme That Model!: A Trusted ML Model Trading Protocol


Title	Gimme That Model!: A Trusted ML Model Trading Protocol
Authors	Laia Amorós, Syed Mahbub Hafiz, Keewoo Lee, M. Caner Tol
Abstract	We propose a HE-based protocol for trading ML models and describe possible improvements to the protocol to make the overall transaction more efficient and secure.
Tasks
Published	2020-03-01
URL	https://arxiv.org/abs/2003.00610v2
PDF	https://arxiv.org/pdf/2003.00610v2.pdf
PWC	https://paperswithcode.com/paper/gimme-that-model-a-trusted-ml-model-trading
Repo
Framework

Constrained Dominant sets and Its applications in computer vision


Title	Constrained Dominant sets and Its applications in computer vision
Authors	Alemu Leulseged Tesfaye
Abstract	In this thesis, we present new schemes which leverage a constrained clustering method to solve several computer vision tasks ranging from image retrieval, image segmentation and co-segmentation, to person re-identification. In the last decades clustering methods have played a vital role in computer vision applications; herein, we focus on the extension, reformulation, and integration of a well-known graph and game theoretic clustering method known as Dominant Sets. Thus, we have demonstrated the validity of the proposed methods with extensive experiments which are conducted on several benchmark datasets.
Tasks	Image Retrieval, Person Re-Identification, Semantic Segmentation
Published	2020-02-12
URL	https://arxiv.org/abs/2002.06028v1
PDF	https://arxiv.org/pdf/2002.06028v1.pdf
PWC	https://paperswithcode.com/paper/constrained-dominant-sets-and-its
Repo
Framework

Real-Time Semantic Background Subtraction


Title	Real-Time Semantic Background Subtraction
Authors	Anthony Cioppa, Marc Van Droogenbroeck, Marc Braham
Abstract	Semantic background subtraction SBS has been shown to improve the performance of most background subtraction algorithms by combining them with semantic information, derived from a semantic segmentation network. However, SBS requires high-quality semantic segmentation masks for all frames, which are slow to compute. In addition, most state-of-the-art background subtraction algorithms are not real-time, which makes them unsuitable for real-world applications. In this paper, we present a novel background subtraction algorithm called Real-Time Semantic Background Subtraction (denoted RT-SBS) which extends SBS for real-time constrained applications while keeping similar performances. RT-SBS effectively combines a real-time background subtraction algorithm with high-quality semantic information which can be provided at a slower pace, independently for each pixel. We show that RT-SBS coupled with ViBe sets a new state of the art for real-time background subtraction algorithms and even competes with the non real-time state-of-the-art ones. Note that python CPU and GPU implementations of RT-SBS will be released soon.
Tasks	Semantic Segmentation
Published	2020-02-12
URL	https://arxiv.org/abs/2002.04993v1
PDF	https://arxiv.org/pdf/2002.04993v1.pdf
PWC	https://paperswithcode.com/paper/real-time-semantic-background-subtraction
Repo
Framework

Adjusting Image Attributes of Localized Regions with Low-level Dialogue


Title	Adjusting Image Attributes of Localized Regions with Low-level Dialogue
Authors	Tzu-Hsiang Lin, Alexander Rudnicky, Trung Bui, Doo Soon Kim, Jean Oh
Abstract	Natural Language Image Editing (NLIE) aims to use natural language instructions to edit images. Since novices are inexperienced with image editing techniques, their instructions are often ambiguous and contain high-level abstractions that tend to correspond to complex editing steps to accomplish. Motivated by this inexperience aspect, we aim to smooth the learning curve by teaching the novices to edit images using low-level commanding terminologies. Towards this end, we develop a task-oriented dialogue system to investigate low-level instructions for NLIE. Our system grounds language on the level of edit operations, and suggests options for a user to choose from. Though compelled to express in low-level terms, a user evaluation shows that 25% of users found our system easy-to-use, resonating with our motivation. An analysis shows that users generally adapt to utilizing the proposed low-level language interface. In this study, we identify that object segmentation as the key factor to the user satisfaction. Our work demonstrates the advantages of the low-level, direct language-action mapping approach that can be applied to other problem domains beyond image editing such as audio editing or industrial design.
Tasks	Semantic Segmentation
Published	2020-02-11
URL	https://arxiv.org/abs/2002.04678v1
PDF	https://arxiv.org/pdf/2002.04678v1.pdf
PWC	https://paperswithcode.com/paper/adjusting-image-attributes-of-localized
Repo
Framework

Explainable Subgraphs with Surprising Densities: A Subgroup Discovery Approach


Title	Explainable Subgraphs with Surprising Densities: A Subgroup Discovery Approach
Authors	Junning Deng, Bo Kang, Jefrey Lijffijt, Tijl De Bie
Abstract	The connectivity structure of graphs is typically related to the attributes of the nodes. In social networks for example, the probability of a friendship between two people depends on their attributes, such as their age, address, and hobbies. The connectivity of a graph can thus possibly be understood in terms of patterns of the form ‘the subgroup of individuals with properties X are often (or rarely) friends with individuals in another subgroup with properties Y’. Such rules present potentially actionable and generalizable insights into the graph. We present a method that finds pairs of node subgroups between which the edge density is interestingly high or low, using an information-theoretic definition of interestingness. This interestingness is quantified subjectively, to contrast with prior information an analyst may have about the graph. This view immediately enables iterative mining of such patterns. Our work generalizes prior work on dense subgraph mining (i.e. subgraphs induced by a single subgroup). Moreover, not only is the proposed method more general, we also demonstrate considerable practical advantages for the single subgroup special case.
Tasks
Published	2020-01-10
URL	https://arxiv.org/abs/2002.00793v1
PDF	https://arxiv.org/pdf/2002.00793v1.pdf
PWC	https://paperswithcode.com/paper/explainable-subgraphs-with-surprising
Repo
Framework

TxSim:Modeling Training of Deep Neural Networks on Resistive Crossbar Systems


Title	TxSim:Modeling Training of Deep Neural Networks on Resistive Crossbar Systems
Authors	Sourjya Roy, Shrihari Sridharan, Shubham Jain, Anand Raghunathan
Abstract	Resistive crossbars have attracted significant interest in the design of Deep Neural Network (DNN) accelerators due to their ability to natively execute massively parallel vector-matrix multiplications within dense memory arrays. However, crossbar-based computations face a major challenge due to a variety of device and circuit-level non-idealities, which manifest as errors in the vector-matrix multiplications and eventually degrade DNN accuracy. To address this challenge, there is a need for tools that can model the functional impact of non-idealities on DNN training and inference. Existing efforts towards this goal are either limited to inference, or are too slow to be used for large-scale DNN training. We propose TxSim, a fast and customizable modeling framework to functionally evaluate DNN training on crossbar-based hardware considering the impact of non-idealities. The key features of TxSim that differentiate it from prior efforts are: (i) It comprehensively models non-idealities during all training operations (forward propagation, backward propagation, and weight update) and (ii) it achieves computational efficiency by mapping crossbar evaluations to well-optimized BLAS routines and incorporates speedup techniques to further reduce simulation time with minimal impact on accuracy. TxSim achieves orders-of-magnitude improvement in simulation speed over prior works, and thereby makes it feasible to evaluate training of large-scale DNNs on crossbars. Our experiments using TxSim reveal that the accuracy degradation in DNN training due to non-idealities can be substantial (3%-10%) for large-scale DNNs, underscoring the need for further research in mitigation techniques. We also analyze the impact of various device and circuit-level parameters and the associated non-idealities to provide key insights that can guide the design of crossbar-based DNN training accelerators.
Tasks
Published	2020-02-25
URL	https://arxiv.org/abs/2002.11151v1
PDF	https://arxiv.org/pdf/2002.11151v1.pdf
PWC	https://paperswithcode.com/paper/txsimmodeling-training-of-deep-neural
Repo
Framework

Document Ranking with a Pretrained Sequence-to-Sequence Model


Title	Document Ranking with a Pretrained Sequence-to-Sequence Model
Authors	Rodrigo Nogueira, Zhiying Jiang, Jimmy Lin
Abstract	This work proposes a novel adaptation of a pretrained sequence-to-sequence model to the task of document ranking. Our approach is fundamentally different from a commonly-adopted classification-based formulation of ranking, based on encoder-only pretrained transformer architectures such as BERT. We show how a sequence-to-sequence model can be trained to generate relevance labels as “target words”, and how the underlying logits of these target words can be interpreted as relevance probabilities for ranking. On the popular MS MARCO passage ranking task, experimental results show that our approach is at least on par with previous classification-based models and can surpass them with larger, more-recent models. On the test collection from the TREC 2004 Robust Track, we demonstrate a zero-shot transfer-based approach that outperforms previous state-of-the-art models requiring in-dataset cross-validation. Furthermore, we find that our approach significantly outperforms an encoder-only model in a data-poor regime (i.e., with few training examples). We investigate this observation further by varying target words to probe the model’s use of latent knowledge.
Tasks	Document Ranking
Published	2020-03-14
URL	https://arxiv.org/abs/2003.06713v1
PDF	https://arxiv.org/pdf/2003.06713v1.pdf
PWC	https://paperswithcode.com/paper/document-ranking-with-a-pretrained-sequence
Repo
Framework

Do Public Datasets Assure Unbiased Comparisons for Registration Evaluation?


Title	Do Public Datasets Assure Unbiased Comparisons for Registration Evaluation?
Authors	Jie Luo, Guangshen Ma, Sarah Frisken, Parikshit Juvekar, Nazim Haouchine, Zhe Xu, Yiming Xiao, Alexandra Golby, Patrick Codd, Masashi Sugiyama, William Wells III
Abstract	With the increasing availability of new image registration approaches, an unbiased evaluation is becoming more needed so that clinicians can choose the most suitable approaches for their applications. Current evaluations typically use landmarks in manually annotated datasets. As a result, the quality of annotations is crucial for unbiased comparisons. Even though most data providers claim to have quality control over their datasets, an objective third-party screening can be reassuring for intended users. In this study, we use the variogram to screen the manually annotated landmarks in two datasets used to benchmark registration in image-guided neurosurgeries. The variogram provides an intuitive 2D representation of the spatial characteristics of annotated landmarks. Using variograms, we identified potentially problematic cases and had them examined by experienced radiologists. We found that (1) a small number of annotations may have fiducial localization errors; (2) the landmark distribution for some cases is not ideal to offer fair comparisons. If unresolved, both findings could incur bias in registration evaluation.
Tasks	Image Registration
Published	2020-03-20
URL	https://arxiv.org/abs/2003.09483v1
PDF	https://arxiv.org/pdf/2003.09483v1.pdf
PWC	https://paperswithcode.com/paper/do-public-datasets-assure-unbiased
Repo
Framework

Abstractive Text Summarization based on Language Model Conditioning and Locality Modeling


Title	Abstractive Text Summarization based on Language Model Conditioning and Locality Modeling
Authors	Dmitrii Aksenov, Julián Moreno-Schneider, Peter Bourgonje, Robert Schwarzenberg, Leonhard Hennig, Georg Rehm
Abstract	We explore to what extent knowledge about the pre-trained language model that is used is beneficial for the task of abstractive summarization. To this end, we experiment with conditioning the encoder and decoder of a Transformer-based neural model on the BERT language model. In addition, we propose a new method of BERT-windowing, which allows chunk-wise processing of texts longer than the BERT window size. We also explore how locality modelling, i.e., the explicit restriction of calculations to the local context, can affect the summarization ability of the Transformer. This is done by introducing 2-dimensional convolutional self-attention into the first layers of the encoder. The results of our models are compared to a baseline and the state-of-the-art models on the CNN/Daily Mail dataset. We additionally train our model on the SwissText dataset to demonstrate usability on German. Both models outperform the baseline in ROUGE scores on two datasets and show its superiority in a manual qualitative analysis.
Tasks	Abstractive Text Summarization, Language Modelling, Text Summarization
Published	2020-03-29
URL	https://arxiv.org/abs/2003.13027v1
PDF	https://arxiv.org/pdf/2003.13027v1.pdf
PWC	https://paperswithcode.com/paper/abstractive-text-summarization-based-on
Repo
Framework