Paper Group ANR 132
Complex spectrogram enhancement by convolutional neural network with multi-metrics learning. Interventions over Predictions: Reframing the Ethical Debate for Actuarial Risk Assessment. Computing the Shapley Value in Allocation Problems: Approximations and Bounds, with an Application to the Italian VQR Research Assessment Program. Speeding up Contex …
Complex spectrogram enhancement by convolutional neural network with multi-metrics learning
| Title | Complex spectrogram enhancement by convolutional neural network with multi-metrics learning |
| Authors | Szu-Wei Fu, Ting-yao Hu, Yu Tsao, Xugang Lu |
| Abstract | This paper aims to address two issues existing in the current speech enhancement methods: 1) the difficulty of phase estimations; 2) a single objective function cannot consider multiple metrics simultaneously. To solve the first problem, we propose a novel convolutional neural network (CNN) model for complex spectrogram enhancement, namely estimating clean real and imaginary (RI) spectrograms from noisy ones. The reconstructed RI spectrograms are directly used to synthesize enhanced speech waveforms. In addition, since log-power spectrogram (LPS) can be represented as a function of RI spectrograms, its reconstruction is also considered as another target. Thus a unified objective function, which combines these two targets (reconstruction of RI spectrograms and LPS), is equivalent to simultaneously optimizing two commonly used objective metrics: segmental signal-to-noise ratio (SSNR) and logspectral distortion (LSD). Therefore, the learning process is called multi-metrics learning (MML). Experimental results confirm the effectiveness of the proposed CNN with RI spectrograms and MML in terms of improved standardized evaluation metrics on a speech enhancement task. |
| Tasks | Speech Enhancement |
| Published | 2017-04-27 |
| URL | http://arxiv.org/abs/1704.08504v2 |
| http://arxiv.org/pdf/1704.08504v2.pdf | |
| PWC | https://paperswithcode.com/paper/complex-spectrogram-enhancement-by |
| Repo | |
| Framework | |
Interventions over Predictions: Reframing the Ethical Debate for Actuarial Risk Assessment
| Title | Interventions over Predictions: Reframing the Ethical Debate for Actuarial Risk Assessment |
| Authors | Chelsea Barabas, Karthik Dinakar, Joichi Ito, Madars Virza, Jonathan Zittrain |
| Abstract | Actuarial risk assessments might be unduly perceived as a neutral way to counteract implicit bias and increase the fairness of decisions made at almost every juncture of the criminal justice system, from pretrial release to sentencing, parole and probation. In recent times these assessments have come under increased scrutiny, as critics claim that the statistical techniques underlying them might reproduce existing patterns of discrimination and historical biases that are reflected in the data. Much of this debate is centered around competing notions of fairness and predictive accuracy, resting on the contested use of variables that act as “proxies” for characteristics legally protected against discrimination, such as race and gender. We argue that a core ethical debate surrounding the use of regression in risk assessments is not simply one of bias or accuracy. Rather, it’s one of purpose. If machine learning is operationalized merely in the service of predicting individual future crime, then it becomes difficult to break cycles of criminalization that are driven by the iatrogenic effects of the criminal justice system itself. We posit that machine learning should not be used for prediction, but rather to surface covariates that are fed into a causal model for understanding the social, structural and psychological drivers of crime. We propose an alternative application of machine learning and causal inference away from predicting risk scores to risk mitigation. |
| Tasks | Causal Inference |
| Published | 2017-12-21 |
| URL | http://arxiv.org/abs/1712.08238v2 |
| http://arxiv.org/pdf/1712.08238v2.pdf | |
| PWC | https://paperswithcode.com/paper/interventions-over-predictions-reframing-the |
| Repo | |
| Framework | |
Computing the Shapley Value in Allocation Problems: Approximations and Bounds, with an Application to the Italian VQR Research Assessment Program
| Title | Computing the Shapley Value in Allocation Problems: Approximations and Bounds, with an Application to the Italian VQR Research Assessment Program |
| Authors | Francesco Lupia, Angelo Mendicelli, Andrea Ribichini, Francesco Scarcello, Marco Schaerf |
| Abstract | In allocation problems, a given set of goods are assigned to agents in such a way that the social welfare is maximised, that is, the largest possible global worth is achieved. When goods are indivisible, it is possible to use money compensation to perform a fair allocation taking into account the actual contribution of all agents to the social welfare. Coalitional games provide a formal mathematical framework to model such problems, in particular the Shapley value is a solution concept widely used for assigning worths to agents in a fair way. Unfortunately, computing this value is a $#{\rm P}$-hard problem, so that applying this good theoretical notion is often quite difficult in real-world problems. We describe useful properties that allow us to greatly simplify the instances of allocation problems, without affecting the Shapley value of any player. Moreover, we propose algorithms for computing lower bounds and upper bounds of the Shapley value, which in some cases provide the exact result and that can be combined with approximation algorithms. The proposed techniques have been implemented and tested on a real-world application of allocation problems, namely, the Italian research assessment program, known as VQR. For the large university considered in the experiments, the problem involves thousands of agents and goods (here, researchers and their research products). The algorithms described in the paper are able to compute the Shapley value for most of those agents, and to get a good approximation of the Shapley value for all of them. |
| Tasks | |
| Published | 2017-09-13 |
| URL | http://arxiv.org/abs/1709.04176v1 |
| http://arxiv.org/pdf/1709.04176v1.pdf | |
| PWC | https://paperswithcode.com/paper/computing-the-shapley-value-in-allocation |
| Repo | |
| Framework | |
Speeding up Context-based Sentence Representation Learning with Non-autoregressive Convolutional Decoding
| Title | Speeding up Context-based Sentence Representation Learning with Non-autoregressive Convolutional Decoding |
| Authors | Shuai Tang, Hailin Jin, Chen Fang, Zhaowen Wang, Virginia R. de Sa |
| Abstract | Context plays an important role in human language understanding, thus it may also be useful for machines learning vector representations of language. In this paper, we explore an asymmetric encoder-decoder structure for unsupervised context-based sentence representation learning. We carefully designed experiments to show that neither an autoregressive decoder nor an RNN decoder is required. After that, we designed a model which still keeps an RNN as the encoder, while using a non-autoregressive convolutional decoder. We further combine a suite of effective designs to significantly improve model efficiency while also achieving better performance. Our model is trained on two different large unlabelled corpora, and in both cases the transferability is evaluated on a set of downstream NLP tasks. We empirically show that our model is simple and fast while producing rich sentence representations that excel in downstream tasks. |
| Tasks | Representation Learning |
| Published | 2017-10-28 |
| URL | http://arxiv.org/abs/1710.10380v3 |
| http://arxiv.org/pdf/1710.10380v3.pdf | |
| PWC | https://paperswithcode.com/paper/speeding-up-context-based-sentence |
| Repo | |
| Framework | |
Data clustering with edge domination in complex networks
| Title | Data clustering with edge domination in complex networks |
| Authors | Paulo Roberto Urio, Zhao Liang |
| Abstract | This paper presents a model for a dynamical system where particles dominate edges in a complex network. The proposed dynamical system is then extended to an application on the problem of community detection and data clustering. In the case of the data clustering problem, 6 different techniques were simulated on 10 different datasets in order to compare with the proposed technique. The results show that the proposed algorithm performs well when prior knowledge of the number of clusters is known to the algorithm. |
| Tasks | Community Detection |
| Published | 2017-05-16 |
| URL | http://arxiv.org/abs/1705.05494v1 |
| http://arxiv.org/pdf/1705.05494v1.pdf | |
| PWC | https://paperswithcode.com/paper/data-clustering-with-edge-domination-in |
| Repo | |
| Framework | |
Partitioning Relational Matrices of Similarities or Dissimilarities using the Value of Information
| Title | Partitioning Relational Matrices of Similarities or Dissimilarities using the Value of Information |
| Authors | Isaac J. Sledge, Jose C. Principe |
| Abstract | In this paper, we provide an approach to clustering relational matrices whose entries correspond to either similarities or dissimilarities between objects. Our approach is based on the value of information, a parameterized, information-theoretic criterion that measures the change in costs associated with changes in information. Optimizing the value of information yields a deterministic annealing style of clustering with many benefits. For instance, investigators avoid needing to a priori specify the number of clusters, as the partitions naturally undergo phase changes, during the annealing process, whereby the number of clusters changes in a data-driven fashion. The global-best partition can also often be identified. |
| Tasks | |
| Published | 2017-10-28 |
| URL | http://arxiv.org/abs/1710.10381v1 |
| http://arxiv.org/pdf/1710.10381v1.pdf | |
| PWC | https://paperswithcode.com/paper/partitioning-relational-matrices-of |
| Repo | |
| Framework | |
ReabsNet: Detecting and Revising Adversarial Examples
| Title | ReabsNet: Detecting and Revising Adversarial Examples |
| Authors | Jiefeng Chen, Zihang Meng, Changtian Sun, Wei Tang, Yinglun Zhu |
| Abstract | Though deep neural network has hit a huge success in recent studies and applica- tions, it still remains vulnerable to adversarial perturbations which are imperceptible to humans. To address this problem, we propose a novel network called ReabsNet to achieve high classification accuracy in the face of various attacks. The approach is to augment an existing classification network with a guardian network to detect if a sample is natural or has been adversarially perturbed. Critically, instead of simply rejecting adversarial examples, we revise them to get their true labels. We exploit the observation that a sample containing adversarial perturbations has a possibility of returning to its true class after revision. We demonstrate that our ReabsNet outperforms the state-of-the-art defense method under various adversarial attacks. |
| Tasks | |
| Published | 2017-12-21 |
| URL | http://arxiv.org/abs/1712.08250v1 |
| http://arxiv.org/pdf/1712.08250v1.pdf | |
| PWC | https://paperswithcode.com/paper/reabsnet-detecting-and-revising-adversarial |
| Repo | |
| Framework | |
Understanding State Preferences With Text As Data: Introducing the UN General Debate Corpus
| Title | Understanding State Preferences With Text As Data: Introducing the UN General Debate Corpus |
| Authors | Alexander Baturo, Niheer Dasandi, Slava J. Mikhaylov |
| Abstract | Every year at the United Nations, member states deliver statements during the General Debate discussing major issues in world politics. These speeches provide invaluable information on governments’ perspectives and preferences on a wide range of issues, but have largely been overlooked in the study of international politics. This paper introduces a new dataset consisting of over 7,701 English-language country statements from 1970-2016. We demonstrate how the UN General Debate Corpus (UNGDC) can be used to derive country positions on different policy dimensions using text analytic methods. The paper provides applications of these estimates, demonstrating the contribution the UNGDC can make to the study of international politics. |
| Tasks | |
| Published | 2017-07-10 |
| URL | http://arxiv.org/abs/1707.02774v1 |
| http://arxiv.org/pdf/1707.02774v1.pdf | |
| PWC | https://paperswithcode.com/paper/understanding-state-preferences-with-text-as |
| Repo | |
| Framework | |
Max-Cosine Matching Based Neural Models for Recognizing Textual Entailment
| Title | Max-Cosine Matching Based Neural Models for Recognizing Textual Entailment |
| Authors | Zhipeng Xie, Junfeng Hu |
| Abstract | Recognizing textual entailment is a fundamental task in a variety of text mining or natural language processing applications. This paper proposes a simple neural model for RTE problem. It first matches each word in the hypothesis with its most-similar word in the premise, producing an augmented representation of the hypothesis conditioned on the premise as a sequence of word pairs. The LSTM model is then used to model this augmented sequence, and the final output from the LSTM is fed into a softmax layer to make the prediction. Besides the base model, in order to enhance its performance, we also proposed three techniques: the integration of multiple word-embedding library, bi-way integration, and ensemble based on model averaging. Experimental results on the SNLI dataset have shown that the three techniques are effective in boosting the predicative accuracy and that our method outperforms several state-of-the-state ones. |
| Tasks | Natural Language Inference |
| Published | 2017-05-25 |
| URL | http://arxiv.org/abs/1705.09054v1 |
| http://arxiv.org/pdf/1705.09054v1.pdf | |
| PWC | https://paperswithcode.com/paper/max-cosine-matching-based-neural-models-for |
| Repo | |
| Framework | |
Rate Optimal Estimation and Confidence Intervals for High-dimensional Regression with Missing Covariates
| Title | Rate Optimal Estimation and Confidence Intervals for High-dimensional Regression with Missing Covariates |
| Authors | Yining Wang, Jialei Wang, Sivaraman Balakrishnan, Aarti Singh |
| Abstract | Although a majority of the theoretical literature in high-dimensional statistics has focused on settings which involve fully-observed data, settings with missing values and corruptions are common in practice. We consider the problems of estimation and of constructing component-wise confidence intervals in a sparse high-dimensional linear regression model when some covariates of the design matrix are missing completely at random. We analyze a variant of the Dantzig selector [9] for estimating the regression model and we use a de-biasing argument to construct component-wise confidence intervals. Our first main result is to establish upper bounds on the estimation error as a function of the model parameters (the sparsity level s, the expected fraction of observed covariates $\rho_*$, and a measure of the signal strength $\beta^*_2$). We find that even in an idealized setting where the covariates are assumed to be missing completely at random, somewhat surprisingly and in contrast to the fully-observed setting, there is a dichotomy in the dependence on model parameters and much faster rates are obtained if the covariance matrix of the random design is known. To study this issue further, our second main contribution is to provide lower bounds on the estimation error showing that this discrepancy in rates is unavoidable in a minimax sense. We then consider the problem of high-dimensional inference in the presence of missing data. We construct and analyze confidence intervals using a de-biased estimator. In the presence of missing data, inference is complicated by the fact that the de-biasing matrix is correlated with the pilot estimator and this necessitates the design of a new estimator and a novel analysis. We also complement our mathematical study with extensive simulations on synthetic and semi-synthetic data that show the accuracy of our asymptotic predictions for finite sample sizes. |
| Tasks | |
| Published | 2017-02-09 |
| URL | http://arxiv.org/abs/1702.02686v2 |
| http://arxiv.org/pdf/1702.02686v2.pdf | |
| PWC | https://paperswithcode.com/paper/rate-optimal-estimation-and-confidence |
| Repo | |
| Framework | |
Conflict management in information fusion with belief functions
| Title | Conflict management in information fusion with belief functions |
| Authors | Arnaud Martin |
| Abstract | In Information fusion, the conflict is an important concept. Indeed, combining several imperfect experts or sources allows conflict. In the theory of belief functions, this notion has been discussed a lot. The mass appearing on the empty set during the conjunctive combination rule is generally considered as conflict, but that is not really a conflict. Some measures of conflict have been proposed and some approaches have been proposed in order to manage this conflict or to decide with conflicting mass functions. We recall in this chapter some of them and we propose a discussion to consider the conflict in information fusion with the theory of belief functions. |
| Tasks | |
| Published | 2017-09-13 |
| URL | http://arxiv.org/abs/1709.04182v1 |
| http://arxiv.org/pdf/1709.04182v1.pdf | |
| PWC | https://paperswithcode.com/paper/conflict-management-in-information-fusion |
| Repo | |
| Framework | |
Understanding Career Progression in Baseball Through Machine Learning
| Title | Understanding Career Progression in Baseball Through Machine Learning |
| Authors | Brian Bierig, Jonathan Hollenbeck, Alexander Stroud |
| Abstract | Professional baseball players are increasingly guaranteed expensive long-term contracts, with over 70 deals signed in excess of $90 million, mostly in the last decade. These are substantial sums compared to a typical franchise valuation of $1-2 billion. Hence, the players to whom a team chooses to give such a contract can have an enormous impact on both competitiveness and profit. Despite this, most published approaches examining career progression in baseball are fairly simplistic. We applied four machine learning algorithms to the problem and soundly improved upon existing approaches, particularly for batting data. |
| Tasks | |
| Published | 2017-12-15 |
| URL | http://arxiv.org/abs/1712.05754v1 |
| http://arxiv.org/pdf/1712.05754v1.pdf | |
| PWC | https://paperswithcode.com/paper/understanding-career-progression-in-baseball |
| Repo | |
| Framework | |
Simultaneous Stereo Video Deblurring and Scene Flow Estimation
| Title | Simultaneous Stereo Video Deblurring and Scene Flow Estimation |
| Authors | Liyuan Pan, Yuchao Dai, Miaomiao Liu, Fatih Porikli |
| Abstract | Videos for outdoor scene often show unpleasant blur effects due to the large relative motion between the camera and the dynamic objects and large depth variations. Existing works typically focus monocular video deblurring. In this paper, we propose a novel approach to deblurring from stereo videos. In particular, we exploit the piece-wise planar assumption about the scene and leverage the scene flow information to deblur the image. Unlike the existing approach [31] which used a pre-computed scene flow, we propose a single framework to jointly estimate the scene flow and deblur the image, where the motion cues from scene flow estimation and blur information could reinforce each other, and produce superior results than the conventional scene flow estimation or stereo deblurring methods. We evaluate our method extensively on two available datasets and achieve significant improvement in flow estimation and removing the blur effect over the state-of-the-art methods. |
| Tasks | Deblurring, Scene Flow Estimation |
| Published | 2017-04-11 |
| URL | http://arxiv.org/abs/1704.03273v1 |
| http://arxiv.org/pdf/1704.03273v1.pdf | |
| PWC | https://paperswithcode.com/paper/simultaneous-stereo-video-deblurring-and |
| Repo | |
| Framework | |
Representation Learning on Large and Small Data
| Title | Representation Learning on Large and Small Data |
| Authors | Chun-Nan Chou, Chuen-Kai Shie, Fu-Chieh Chang, Jocelyn Chang, Edward Y. Chang |
| Abstract | Deep learning owes its success to three key factors: scale of data, enhanced models to learn representations from data, and scale of computation. This book chapter presented the importance of the data-driven approach to learn good representations from both big data and small data. In terms of big data, it has been widely accepted in the research community that the more data the better for both representation and classification improvement. The question is then how to learn representations from big data, and how to perform representation learning when data is scarce. We addressed the first question by presenting CNN model enhancements in the aspects of representation, optimization, and generalization. To address the small data challenge, we showed transfer representation learning to be effective. Transfer representation learning transfers the learned representation from a source domain where abundant training data is available to a target domain where training data is scarce. Transfer representation learning gave the OM and melanoma diagnosis modules of our XPRIZE Tricorder device (which finished $2^{nd}$ out of $310$ competing teams) a significant boost in diagnosis accuracy. |
| Tasks | Representation Learning |
| Published | 2017-07-25 |
| URL | http://arxiv.org/abs/1707.09873v1 |
| http://arxiv.org/pdf/1707.09873v1.pdf | |
| PWC | https://paperswithcode.com/paper/representation-learning-on-large-and-small |
| Repo | |
| Framework | |
The cost of fairness in classification
| Title | The cost of fairness in classification |
| Authors | Aditya Krishna Menon, Robert C. Williamson |
| Abstract | We study the problem of learning classifiers with a fairness constraint, with three main contributions towards the goal of quantifying the problem’s inherent tradeoffs. First, we relate two existing fairness measures to cost-sensitive risks. Second, we show that for cost-sensitive classification and fairness measures, the optimal classifier is an instance-dependent thresholding of the class-probability function. Third, we show how the tradeoff between accuracy and fairness is determined by the alignment between the class-probabilities for the target and sensitive features. Underpinning our analysis is a general framework that casts the problem of learning with a fairness requirement as one of minimising the difference of two statistical risks. |
| Tasks | |
| Published | 2017-05-25 |
| URL | http://arxiv.org/abs/1705.09055v1 |
| http://arxiv.org/pdf/1705.09055v1.pdf | |
| PWC | https://paperswithcode.com/paper/the-cost-of-fairness-in-classification |
| Repo | |
| Framework | |
