July 29, 2019

3244 words 16 mins read

Paper Group ANR 82

Recurrent Attentional Reinforcement Learning for Multi-label Image Recognition. Evidence combination for a large number of sources. Preference-based Teaching. A Denoising Loss Bound for Neural Network based Universal Discrete Denoisers. Fairer and more accurate, but for whom?. Finding Dominant User Utterances And System Responses in Conversations. …

Recurrent Attentional Reinforcement Learning for Multi-label Image Recognition


Title	Recurrent Attentional Reinforcement Learning for Multi-label Image Recognition
Authors	Tianshui Chen, Zhouxia Wang, Guanbin Li, Liang Lin
Abstract	Recognizing multiple labels of images is a fundamental but challenging task in computer vision, and remarkable progress has been attained by localizing semantic-aware image regions and predicting their labels with deep convolutional neural networks. The step of hypothesis regions (region proposals) localization in these existing multi-label image recognition pipelines, however, usually takes redundant computation cost, e.g., generating hundreds of meaningless proposals with non-discriminative information and extracting their features, and the spatial contextual dependency modeling among the localized regions are often ignored or over-simplified. To resolve these issues, this paper proposes a recurrent attention reinforcement learning framework to iteratively discover a sequence of attentional and informative regions that are related to different semantic objects and further predict label scores conditioned on these regions. Besides, our method explicitly models long-term dependencies among these attentional regions that help to capture semantic label co-occurrence and thus facilitate multi-label recognition. Extensive experiments and comparisons on two large-scale benchmarks (i.e., PASCAL VOC and MS-COCO) show that our model achieves superior performance over existing state-of-the-art methods in both performance and efficiency as well as explicitly identifying image-level semantic labels to specific object regions.
Tasks
Published	2017-12-20
URL	http://arxiv.org/abs/1712.07465v1
PDF	http://arxiv.org/pdf/1712.07465v1.pdf
PWC	https://paperswithcode.com/paper/recurrent-attentional-reinforcement-learning
Repo
Framework

Evidence combination for a large number of sources


Title	Evidence combination for a large number of sources
Authors	Kuang Zhou, Arnaud Martin, Quan Pan
Abstract	The theory of belief functions is an effective tool to deal with the multiple uncertain information. In recent years, many evidence combination rules have been proposed in this framework, such as the conjunctive rule, the cautious rule, the PCR (Proportional Conflict Redistribution) rules and so on. These rules can be adopted for different types of sources. However, most of these rules are not applicable when the number of sources is large. This is due to either the complexity or the existence of an absorbing element (such as the total conflict mass function for the conjunctive-based rules when applied on unreliable evidence). In this paper, based on the assumption that the majority of sources are reliable, a combination rule for a large number of sources, named LNS (stands for Large Number of Sources), is proposed on the basis of a simple idea: the more common ideas one source shares with others, the morereliable the source is. This rule is adaptable for aggregating a large number of sources among which some are unreliable. It will keep the spirit of the conjunctive rule to reinforce the belief on the focal elements with which the sources are in agreement. The mass on the empty set will be kept as an indicator of the conflict. Moreover, it can be used to elicit the major opinion among the experts. The experimental results on synthetic mass functionsverify that the rule can be effectively used to combine a large number of mass functions and to elicit the major opinion.
Tasks
Published	2017-07-25
URL	http://arxiv.org/abs/1707.07999v1
PDF	http://arxiv.org/pdf/1707.07999v1.pdf
PWC	https://paperswithcode.com/paper/evidence-combination-for-a-large-number-of
Repo
Framework

Preference-based Teaching


Title	Preference-based Teaching
Authors	Ziyuan Gao, Christoph Ries, Hans Ulrich Simon, Sandra Zilles
Abstract	We introduce a new model of teaching named “preference-based teaching” and a corresponding complexity parameter—the preference-based teaching dimension (PBTD)—representing the worst-case number of examples needed to teach any concept in a given concept class. Although the PBTD coincides with the well-known recursive teaching dimension (RTD) on finite classes, it is radically different on infinite ones: the RTD becomes infinite already for trivial infinite classes (such as half-intervals) whereas the PBTD evaluates to reasonably small values for a wide collection of infinite classes including classes consisting of so-called closed sets w.r.t. a given closure operator, including various classes related to linear sets over $\mathbb{N}_0$ (whose RTD had been studied quite recently) and including the class of Euclidean half-spaces. On top of presenting these concrete results, we provide the reader with a theoretical framework (of a combinatorial flavor) which helps to derive bounds on the PBTD.
Tasks
Published	2017-02-06
URL	http://arxiv.org/abs/1702.02047v2
PDF	http://arxiv.org/pdf/1702.02047v2.pdf
PWC	https://paperswithcode.com/paper/preference-based-teaching
Repo
Framework

A Denoising Loss Bound for Neural Network based Universal Discrete Denoisers


Title	A Denoising Loss Bound for Neural Network based Universal Discrete Denoisers
Authors	Taesup Moon
Abstract	We obtain a denoising loss bound of the recently proposed neural network based universal discrete denoiser, Neural DUDE, which can adaptively learn its parameters solely from the noise-corrupted data, by minimizing the \emph{empirical estimated loss}. The resulting bound resembles the generalization error bound of the standard empirical risk minimizers (ERM) in supervised learning, and we show that the well-known bias-variance tradeoff also exists in our loss bound. The key tool we develop is the concentration of the unbiased estimated loss on the true denoising loss, which is shown to hold \emph{uniformly} for \emph{all} bounded network parameters and \emph{all} underlying clean sequences. For proving our main results, we make a novel application of the tools from the statistical learning theory. Finally, we show that the hyperparameters of Neural DUDE can be chosen from a small validation set to significantly improve the denoising performance, as predicted by the theoretical result of this paper.
Tasks	Denoising
Published	2017-09-12
URL	http://arxiv.org/abs/1709.03657v2
PDF	http://arxiv.org/pdf/1709.03657v2.pdf
PWC	https://paperswithcode.com/paper/a-denoising-loss-bound-for-neural-network
Repo
Framework

Fairer and more accurate, but for whom?


Title	Fairer and more accurate, but for whom?
Authors	Alexandra Chouldechova, Max G’Sell
Abstract	Complex statistical machine learning models are increasingly being used or considered for use in high-stakes decision-making pipelines in domains such as financial services, health care, criminal justice and human services. These models are often investigated as possible improvements over more classical tools such as regression models or human judgement. While the modeling approach may be new, the practice of using some form of risk assessment to inform decisions is not. When determining whether a new model should be adopted, it is therefore essential to be able to compare the proposed model to the existing approach across a range of task-relevant accuracy and fairness metrics. Looking at overall performance metrics, however, may be misleading. Even when two models have comparable overall performance, they may nevertheless disagree in their classifications on a considerable fraction of cases. In this paper we introduce a model comparison framework for automatically identifying subgroups in which the differences between models are most pronounced. Our primary focus is on identifying subgroups where the models differ in terms of fairness-related quantities such as racial or gender disparities. We present experimental results from a recidivism prediction task and a hypothetical lending example.
Tasks	Decision Making
Published	2017-06-30
URL	http://arxiv.org/abs/1707.00046v1
PDF	http://arxiv.org/pdf/1707.00046v1.pdf
PWC	https://paperswithcode.com/paper/fairer-and-more-accurate-but-for-whom
Repo
Framework

Finding Dominant User Utterances And System Responses in Conversations


Title	Finding Dominant User Utterances And System Responses in Conversations
Authors	Dhiraj Madan, Sachindra Joshi
Abstract	There are several dialog frameworks which allow manual specification of intents and rule based dialog flow. The rule based framework provides good control to dialog designers at the expense of being more time consuming and laborious. The job of a dialog designer can be reduced if we could identify pairs of user intents and corresponding responses automatically from prior conversations between users and agents. In this paper we propose an approach to find these frequent user utterances (which serve as examples for intents) and corresponding agent responses. We propose a novel SimCluster algorithm that extends standard K-means algorithm to simultaneously cluster user utterances and agent utterances by taking their adjacency information into account. The method also aligns these clusters to provide pairs of intents and response groups. We compare our results with those produced by using simple Kmeans clustering on a real dataset and observe upto 10% absolute improvement in F1-scores. Through our experiments on synthetic dataset, we show that our algorithm gains more advantage over K-means algorithm when the data has large variance.
Tasks
Published	2017-10-29
URL	http://arxiv.org/abs/1710.10609v1
PDF	http://arxiv.org/pdf/1710.10609v1.pdf
PWC	https://paperswithcode.com/paper/finding-dominant-user-utterances-and-system
Repo
Framework

Differentially Private Distributed Learning for Language Modeling Tasks


Title	Differentially Private Distributed Learning for Language Modeling Tasks
Authors	Vadim Popov, Mikhail Kudinov, Irina Piontkovskaya, Petr Vytovtov, Alex Nevidomsky
Abstract	One of the big challenges in machine learning applications is that training data can be different from the real-world data faced by the algorithm. In language modeling, users’ language (e.g. in private messaging) could change in a year and be completely different from what we observe in publicly available data. At the same time, public data can be used for obtaining general knowledge (i.e. general model of English). We study approaches to distributed fine-tuning of a general model on user private data with the additional requirements of maintaining the quality on the general data and minimization of communication costs. We propose a novel technique that significantly improves prediction quality on users’ language compared to a general model and outperforms gradient compression methods in terms of communication efficiency. The proposed procedure is fast and leads to an almost 70% perplexity reduction and 8.7 percentage point improvement in keystroke saving rate on informal English texts. We also show that the range of tasks our approach is applicable to is not limited by language modeling only. Finally, we propose an experimental framework for evaluating differential privacy of distributed training of language models and show that our approach has good privacy guarantees.
Tasks	Language Modelling
Published	2017-12-20
URL	http://arxiv.org/abs/1712.07473v3
PDF	http://arxiv.org/pdf/1712.07473v3.pdf
PWC	https://paperswithcode.com/paper/differentially-private-distributed-learning
Repo
Framework

Optimal sequential treatment allocation


Title	Optimal sequential treatment allocation
Authors	Anders Bredahl Kock, Martin Thyrsgaard
Abstract	In treatment allocation problems the individuals to be treated often arrive sequentially. We study a problem in which the policy maker is not only interested in the expected cumulative welfare but is also concerned about the uncertainty/risk of the treatment outcomes. At the outset, the total number of treatment assignments to be made may even be unknown. A sequential treatment policy which attains the minimax optimal regret is proposed. We also demonstrate that the expected number of suboptimal treatments only grows slowly in the number of treatments. Finally, we study a setting where outcomes are only observed with delay.
Tasks
Published	2017-05-28
URL	http://arxiv.org/abs/1705.09952v4
PDF	http://arxiv.org/pdf/1705.09952v4.pdf
PWC	https://paperswithcode.com/paper/optimal-sequential-treatment-allocation
Repo
Framework

Deep Head Pose Estimation from Depth Data for In-car Automotive Applications


Title	Deep Head Pose Estimation from Depth Data for In-car Automotive Applications
Authors	Marco Venturelli, Guido Borghi, Roberto Vezzani, Rita Cucchiara
Abstract	Recently, deep learning approaches have achieved promising results in various fields of computer vision. In this paper, we tackle the problem of head pose estimation through a Convolutional Neural Network (CNN). Differently from other proposals in the literature, the described system is able to work directly and based only on raw depth data. Moreover, the head pose estimation is solved as a regression problem and does not rely on visual facial features like facial landmarks. We tested our system on a well known public dataset, Biwi Kinect Head Pose, showing that our approach achieves state-of-art results and is able to meet real time performance requirements.
Tasks	Head Pose Estimation, Pose Estimation
Published	2017-03-06
URL	http://arxiv.org/abs/1703.01883v1
PDF	http://arxiv.org/pdf/1703.01883v1.pdf
PWC	https://paperswithcode.com/paper/deep-head-pose-estimation-from-depth-data-for
Repo
Framework

Scalable Surface Reconstruction from Point Clouds with Extreme Scale and Density Diversity


Title	Scalable Surface Reconstruction from Point Clouds with Extreme Scale and Density Diversity
Authors	Christian Mostegel, Rudolf Prettenthaler, Friedrich Fraundorfer, Horst Bischof
Abstract	In this paper we present a scalable approach for robustly computing a 3D surface mesh from multi-scale multi-view stereo point clouds that can handle extreme jumps of point density (in our experiments three orders of magnitude). The backbone of our approach is a combination of octree data partitioning, local Delaunay tetrahedralization and graph cut optimization. Graph cut optimization is used twice, once to extract surface hypotheses from local Delaunay tetrahedralizations and once to merge overlapping surface hypotheses even when the local tetrahedralizations do not share the same topology.This formulation allows us to obtain a constant memory consumption per sub-problem while at the same time retaining the density independent interpolation properties of the Delaunay-based optimization. On multiple public datasets, we demonstrate that our approach is highly competitive with the state-of-the-art in terms of accuracy, completeness and outlier resilience. Further, we demonstrate the multi-scale potential of our approach by processing a newly recorded dataset with 2 billion points and a point density variation of more than four orders of magnitude - requiring less than 9GB of RAM per process.
Tasks
Published	2017-05-02
URL	http://arxiv.org/abs/1705.00949v1
PDF	http://arxiv.org/pdf/1705.00949v1.pdf
PWC	https://paperswithcode.com/paper/scalable-surface-reconstruction-from-point
Repo
Framework

Deep Learning for Metagenomic Data: using 2D Embeddings and Convolutional Neural Networks


Title	Deep Learning for Metagenomic Data: using 2D Embeddings and Convolutional Neural Networks
Authors	Thanh Hai Nguyen, Yann Chevaleyre, Edi Prifti, Nataliya Sokolovska, Jean-Daniel Zucker
Abstract	Deep learning (DL) techniques have had unprecedented success when applied to images, waveforms, and texts to cite a few. In general, when the sample size (N) is much greater than the number of features (d), DL outperforms previous machine learning (ML) techniques, often through the use of convolution neural networks (CNNs). However, in many bioinformatics ML tasks, we encounter the opposite situation where d is greater than N. In these situations, applying DL techniques (such as feed-forward networks) would lead to severe overfitting. Thus, sparse ML techniques (such as LASSO e.g.) usually yield the best results on these tasks. In this paper, we show how to apply CNNs on data which do not have originally an image structure (in particular on metagenomic data). Our first contribution is to show how to map metagenomic data in a meaningful way to 1D or 2D images. Based on this representation, we then apply a CNN, with the aim of predicting various diseases. The proposed approach is applied on six different datasets including in total over 1000 samples from various diseases. This approach could be a promising one for prediction tasks in the bioinformatics field.
Tasks
Published	2017-12-01
URL	http://arxiv.org/abs/1712.00244v1
PDF	http://arxiv.org/pdf/1712.00244v1.pdf
PWC	https://paperswithcode.com/paper/deep-learning-for-metagenomic-data-using-2d
Repo
Framework

Provably efficient neural network representation for image classification


Title	Provably efficient neural network representation for image classification
Authors	Yichen Huang
Abstract	The state-of-the-art approaches for image classification are based on neural networks. Mathematically, the task of classifying images is equivalent to finding the function that maps an image to the label it is associated with. To rigorously establish the success of neural network methods, we should first prove that the function has an efficient neural network representation, and then design provably efficient training algorithms to find such a representation. Here, we achieve the first goal based on a set of assumptions about the patterns in the images. The validity of these assumptions is very intuitive in many image classification problems, including but not limited to, recognizing handwritten digits.
Tasks	Image Classification
Published	2017-11-13
URL	http://arxiv.org/abs/1711.04606v1
PDF	http://arxiv.org/pdf/1711.04606v1.pdf
PWC	https://paperswithcode.com/paper/provably-efficient-neural-network
Repo
Framework

An Empirical Analysis of NMT-Derived Interlingual Embeddings and their Use in Parallel Sentence Identification


Title	An Empirical Analysis of NMT-Derived Interlingual Embeddings and their Use in Parallel Sentence Identification
Authors	Cristina España-Bonet, Ádám Csaba Varga, Alberto Barrón-Cedeño, Josef van Genabith
Abstract	End-to-end neural machine translation has overtaken statistical machine translation in terms of translation quality for some language pairs, specially those with large amounts of parallel data. Besides this palpable improvement, neural networks provide several new properties. A single system can be trained to translate between many languages at almost no additional cost other than training time. Furthermore, internal representations learned by the network serve as a new semantic representation of words -or sentences- which, unlike standard word embeddings, are learned in an essentially bilingual or even multilingual context. In view of these properties, the contribution of the present work is two-fold. First, we systematically study the NMT context vectors, i.e. output of the encoder, and their power as an interlingua representation of a sentence. We assess their quality and effectiveness by measuring similarities across translations, as well as semantically related and semantically unrelated sentence pairs. Second, as extrinsic evaluation of the first point, we identify parallel sentences in comparable corpora, obtaining an F1=98.2% on data from a shared task when using only NMT context vectors. Using context vectors jointly with similarity measures F1 reaches 98.9%.
Tasks	Machine Translation, Word Embeddings
Published	2017-04-18
URL	http://arxiv.org/abs/1704.05415v2
PDF	http://arxiv.org/pdf/1704.05415v2.pdf
PWC	https://paperswithcode.com/paper/an-empirical-analysis-of-nmt-derived
Repo
Framework

Effective Image Retrieval via Multilinear Multi-index Fusion


Title	Effective Image Retrieval via Multilinear Multi-index Fusion
Authors	Zhizhong Zhang, Yuan Xie, Wensheng Zhang, Qi Tian
Abstract	Multi-index fusion has demonstrated impressive performances in retrieval task by integrating different visual representations in a unified framework. However, previous works mainly consider propagating similarities via neighbor structure, ignoring the high order information among different visual representations. In this paper, we propose a new multi-index fusion scheme for image retrieval. By formulating this procedure as a multilinear based optimization problem, the complementary information hidden in different indexes can be explored more thoroughly. Specially, we first build our multiple indexes from various visual representations. Then a so-called index-specific functional matrix, which aims to propagate similarities, is introduced for updating the original index. The functional matrices are then optimized in a unified tensor space to achieve a refinement, such that the relevant images can be pushed more closer. The optimization problem can be efficiently solved by the augmented Lagrangian method with theoretical convergence guarantee. Unlike the traditional multi-index fusion scheme, our approach embeds the multi-index subspace structure into the new indexes with sparse constraint, thus it has little additional memory consumption in online query stage. Experimental evaluation on three benchmark datasets reveals that the proposed approach achieves the state-of-the-art performance, i.e., N-score 3.94 on UKBench, mAP 94.1% on Holiday and 62.39% on Market-1501.
Tasks	Image Retrieval
Published	2017-09-27
URL	http://arxiv.org/abs/1709.09304v1
PDF	http://arxiv.org/pdf/1709.09304v1.pdf
PWC	https://paperswithcode.com/paper/effective-image-retrieval-via-multilinear
Repo
Framework

Automatic Emphysema Detection using Weakly Labeled HRCT Lung Images


Title	Automatic Emphysema Detection using Weakly Labeled HRCT Lung Images
Authors	Isabel Pino Peña, Veronika Cheplygina, Sofia Paschaloudi, Morten Vuust, Jesper Carl, Ulla Møller Weinreich, Lasse Riis Østergaard, Marleen de Bruijne
Abstract	A method for automatically quantifying emphysema regions using High-Resolution Computed Tomography (HRCT) scans of patients with chronic obstructive pulmonary disease (COPD) that does not require manually annotated scans for training is presented. HRCT scans of controls and of COPD patients with diverse disease severity are acquired at two different centers. Textural features from co-occurrence matrices and Gaussian filter banks are used to characterize the lung parenchyma in the scans. Two robust versions of multiple instance learning (MIL) classifiers, miSVM and MILES, are investigated. The classifiers are trained with the weak labels extracted from the forced expiratory volume in one minute (FEV$_1$) and diffusing capacity of the lungs for carbon monoxide (DLCO). At test time, the classifiers output a patient label indicating overall COPD diagnosis and local labels indicating the presence of emphysema. The classifier performance is compared with manual annotations by two radiologists, a classical density based method, and pulmonary function tests (PFTs). The miSVM classifier performed better than MILES on both patient and emphysema classification. The classifier has a stronger correlation with PFT than the density based method, the percentage of emphysema in the intersection of annotations from both radiologists, and the percentage of emphysema annotated by one of the radiologists. The correlation between the classifier and the PFT is only outperformed by the second radiologist. The method is therefore promising for facilitating assessment of emphysema and reducing inter-observer variability.
Tasks	Multiple Instance Learning
Published	2017-06-07
URL	http://arxiv.org/abs/1706.02051v2
PDF	http://arxiv.org/pdf/1706.02051v2.pdf
PWC	https://paperswithcode.com/paper/automatic-emphysema-detection-using-weakly
Repo
Framework