January 26, 2020

2901 words 14 mins read

Paper Group ANR 1489

On the statistical evaluation of algorithmic’s computational experimentation with infeasible solutions. Orthogonal Nonnegative Tucker Decomposition. Revisiting the Approximate Carathéodory Problem via the Frank-Wolfe Algorithm. Deep neural networks can predict mortality from 12-lead electrocardiogram voltage data. Prediction Modeling and Analysis f …

On the statistical evaluation of algorithmic’s computational experimentation with infeasible solutions


Title	On the statistical evaluation of algorithmic’s computational experimentation with infeasible solutions
Authors	Iago A Carvalho
Abstract	The experimental evaluation of algorithms results in a large set of data which generally do not follow a normal distribution or are not heteroscedastic. Besides, some of its entries may be missing, due to the inability of an algorithm to find a feasible solution until a time limit is met. Those characteristics restrict the statistical evaluation of computational experiments. This work proposes a bi-objective lexicographical ranking scheme to evaluate datasets with such characteristics. The output ranking can be used as input to any desired statistical test. We used the proposed ranking scheme to assess the results obtained by the Iterative Rounding heuristic (IR). A Friedman’s test and a subsequent post-hoc test carried out on the ranked data demonstrated that IR performed significantly better than the Feasibility Pump heuristic when solving 152 benchmark problems of Nonconvex Mixed-Integer Nonlinear Problems. However, is also showed that the RECIPE heuristic was significantly better than IR when solving the same benchmark problems.
Tasks
Published	2019-01-31
URL	http://arxiv.org/abs/1902.00101v1
PDF	http://arxiv.org/pdf/1902.00101v1.pdf
PWC	https://paperswithcode.com/paper/on-the-statistical-evaluation-of-algorithmics
Repo
Framework

Orthogonal Nonnegative Tucker Decomposition


Title	Orthogonal Nonnegative Tucker Decomposition
Authors	Junjun Pan, Michael K. Ng, Ye Liu, Xiongjun Zhang, Hong Yan
Abstract	In this paper, we study the nonnegative tensor data and propose an orthogonal nonnegative Tucker decomposition (ONTD). We discuss some properties of ONTD and develop a convex relaxation algorithm of the augmented Lagrangian function to solve the optimization problem. The convergence of the algorithm is given. We employ ONTD on the image data sets from the real world applications including face recognition, image representation, hyperspectral unmixing. Numerical results are shown to illustrate the effectiveness of the proposed algorithm.
Tasks	Face Recognition, Hyperspectral Unmixing
Published	2019-10-21
URL	https://arxiv.org/abs/1910.09979v2
PDF	https://arxiv.org/pdf/1910.09979v2.pdf
PWC	https://paperswithcode.com/paper/orthogonal-nonnegative-tucker-decomposition
Repo
Framework

Revisiting the Approximate Carathéodory Problem via the Frank-Wolfe Algorithm


Title	Revisiting the Approximate Carathéodory Problem via the Frank-Wolfe Algorithm
Authors	Cyrille W. Combettes, Sebastian Pokutta
Abstract	The approximate Carath'eodory theorem states that given a polytope $\mathcal{P}$, each point in $\mathcal{P}$ can be approximated within $\epsilon$-accuracy in $\ell_p$-norm as the convex combination of $\mathcal{O}(pD_p^2/\epsilon^2)$ vertices, where $p\geq2$ and $D_p$ is the diameter of $\mathcal{P}$ in $\ell_p$-norm. A solution satisfying these properties can be built using probabilistic arguments [Barman, 2015] or by applying mirror descent to the dual problem [Mirrokni et al., 2017]. We revisit the approximate Carath'eodory problem by solving the primal problem via the Frank-Wolfe algorithm, providing a simplified analysis and leading to an efficient practical method. Sublinear to linear sparsity bounds are derived naturally using existing convergence results of the Frank-Wolfe algorithm in different scenarios.
Tasks
Published	2019-11-11
URL	https://arxiv.org/abs/1911.04415v1
PDF	https://arxiv.org/pdf/1911.04415v1.pdf
PWC	https://paperswithcode.com/paper/revisiting-the-approximate-caratheodory
Repo
Framework

Deep neural networks can predict mortality from 12-lead electrocardiogram voltage data


Title	Deep neural networks can predict mortality from 12-lead electrocardiogram voltage data
Authors	Sushravya Raghunath, Alvaro E. Ulloa Cerna, Linyuan Jing, David P. vanMaanen, Joshua Stough, Dustin N. Hartzel, Joseph B. Leader, H. Lester Kirchner, Christopher W. Good, Aalpen A. Patel, Brian P. Delisle, Amro Alsaid, Dominik Beer, Christopher M. Haggerty, Brandon K. Fornwalt
Abstract	The electrocardiogram (ECG) is a widely-used medical test, typically consisting of 12 voltage versus time traces collected from surface recordings over the heart. Here we hypothesize that a deep neural network can predict an important future clinical event (one-year all-cause mortality) from ECG voltage-time traces. We show good performance for predicting one-year mortality with an average AUC of 0.85 from a model cross-validated on 1,775,926 12-lead resting ECGs, that were collected over a 34-year period in a large regional health system. Even within the large subset of ECGs interpreted as ‘normal’ by a physician (n=297,548), the model performance to predict one-year mortality remained high (AUC=0.84), and Cox Proportional Hazard model revealed a hazard ratio of 6.6 (p<0.005) for the two predicted groups (dead vs alive one year after ECG) over a 30-year follow-up period. A blinded survey of three cardiologists suggested that the patterns captured by the model were generally not visually apparent to cardiologists even after being shown 240 paired examples of labeled true positives (dead) and true negatives (alive). In summary, deep learning can add significant prognostic information to the interpretation of 12-lead resting ECGs, even in cases that are interpreted as ‘normal’ by physicians.
Tasks	Electrocardiography (ECG), Mortality Prediction
Published	2019-04-15
URL	https://arxiv.org/abs/1904.07032v2
PDF	https://arxiv.org/pdf/1904.07032v2.pdf
PWC	https://paperswithcode.com/paper/deep-neural-networks-can-predict-mortality
Repo
Framework

Prediction Modeling and Analysis for Telecom Customer Churn in Two Months


Title	Prediction Modeling and Analysis for Telecom Customer Churn in Two Months
Authors	Lingling Yang, Dongyang Li, Yao Lu
Abstract	A practical churn customer prediction model is critical to retain customers for telecom companies in the saturated and competitive market. Previous studies focus on predicting churn customers in current or next month, in which telecom companies don’t have enough time to develop and carry out churn management strategies. In this paper, we propose a new T+2 churn customer prediction model, in which the churn customers in two months are recognized and the one-month window T+1 is reserved to carry out churn management strategies. However, the predictions for churn customers in two months are much more difficult than in current or next month because of the weaker correlation between the customer information and churn states. Two characteristics of telecom dataset, the discrimination between churn and non-churn customers is complicated and the class imbalance problem is serious, are observed. To discriminate the churn customers accurately, random forest (RF) classifier is chosen because RF solves the nonlinear separable problem with low bias and low variance and handles high feature spaces and large number of training examples. To overcome the imbalance problem, synthetic minority over-sampling with borderline or tomek link, in which the distribution of the samples remains and the number of the training examples becomes larger, is applied. Overall, a precision ratio of about 50% with a recall ratio of about 50% is achieved in the T+2 churn prediction. The proposed prediction model provides an accurate and operable churn customer prediction model for telecom companies.
Tasks
Published	2019-11-01
URL	https://arxiv.org/abs/1911.00558v1
PDF	https://arxiv.org/pdf/1911.00558v1.pdf
PWC	https://paperswithcode.com/paper/prediction-modeling-and-analysis-for-telecom
Repo
Framework

Feature selection of neural networks is skewed towards the less abstract cue


Title	Feature selection of neural networks is skewed towards the less abstract cue
Authors	Marcell Wolnitza, Babette Dellen
Abstract	Artificial neural networks (ANNs) have become an important tool for image classification with many applications in research and industry. However, it remains largely unknown how relevant image features are selected and how data properties affect this process. In particular, we are interested whether the abstraction level of image cues correlating with class membership influences feature selection. We perform experiments with binary images that contain a combination of cues, representing two different levels of abstractions: one is a pattern drawn from a random distribution where class membership correlates with the statistics of the pattern, the other a combination of symbol-like entities, where the symbolic code correlates with class membership. When the network is trained with data in which both cues are equally significant, we observe that the cues at the lower abstraction level, i.e., the pattern, is learned, while the symbolic information is largely ignored, even in networks with many layers. Symbol-like entities are only learned if the importance of low-level cues is reduced compared to the high-level ones. These findings raise important questions about the relevance of features that are learned by deep ANNs and how learning could be shifted towards symbolic features.
Tasks	Feature Selection, Image Classification
Published	2019-08-08
URL	https://arxiv.org/abs/1908.03000v1
PDF	https://arxiv.org/pdf/1908.03000v1.pdf
PWC	https://paperswithcode.com/paper/feature-selection-of-neural-networks-is
Repo
Framework

Differential Bayesian Neural Nets


Title	Differential Bayesian Neural Nets
Authors	Andreas Look, Melih Kandemir
Abstract	Neural Ordinary Differential Equations (N-ODEs) are a powerful building block for learning systems, which extend residual networks to a continuous-time dynamical system. We propose a Bayesian version of N-ODEs that enables well-calibrated quantification of prediction uncertainty, while maintaining the expressive power of their deterministic counterpart. We assign Bayesian Neural Nets (BNNs) to both the drift and the diffusion terms of a Stochastic Differential Equation (SDE) that models the flow of the activation map in time. We infer the posterior on the BNN weights using a straightforward adaptation of Stochastic Gradient Langevin Dynamics (SGLD). We illustrate significantly improved stability on two synthetic time series prediction tasks and report better model fit on UCI regression benchmarks with our method when compared to its non-Bayesian counterpart.
Tasks	Time Series, Time Series Prediction
Published	2019-12-02
URL	https://arxiv.org/abs/1912.00796v2
PDF	https://arxiv.org/pdf/1912.00796v2.pdf
PWC	https://paperswithcode.com/paper/differential-bayesian-neural-nets
Repo
Framework

Certified Robustness to Adversarial Word Substitutions


Title	Certified Robustness to Adversarial Word Substitutions
Authors	Robin Jia, Aditi Raghunathan, Kerem Göksel, Percy Liang
Abstract	State-of-the-art NLP models can often be fooled by adversaries that apply seemingly innocuous label-preserving transformations (e.g., paraphrasing) to input text. The number of possible transformations scales exponentially with text length, so data augmentation cannot cover all transformations of an input. This paper considers one exponentially large family of label-preserving transformations, in which every word in the input can be replaced with a similar word. We train the first models that are provably robust to all word substitutions in this family. Our training procedure uses Interval Bound Propagation (IBP) to minimize an upper bound on the worst-case loss that any combination of word substitutions can induce. To evaluate models’ robustness to these transformations, we measure accuracy on adversarially chosen word substitutions applied to test examples. Our IBP-trained models attain $75%$ adversarial accuracy on both sentiment analysis on IMDB and natural language inference on SNLI. In comparison, on IMDB, models trained normally and ones trained with data augmentation achieve adversarial accuracy of only $8%$ and $35%$, respectively.
Tasks	Data Augmentation, Natural Language Inference, Sentiment Analysis
Published	2019-09-03
URL	https://arxiv.org/abs/1909.00986v1
PDF	https://arxiv.org/pdf/1909.00986v1.pdf
PWC	https://paperswithcode.com/paper/certified-robustness-to-adversarial-word
Repo
Framework

Maximum likelihood estimation for disk image parameters


Title	Maximum likelihood estimation for disk image parameters
Authors	Matwey V. Kornilov
Abstract	We present a novel technique for estimating disk parameters (the centre and the radius) from its 2D image. It is based on the maximal likelihood approach utilising both edge pixels coordinates and the image intensity gradients. We emphasise the following advantages of our likelihood model. It has closed-form formulae for parameter estimating, requiring less computational resources than iterative algorithms therefore. The likelihood model naturally distinguishes the outer and inner annulus edges. The proposed technique was evaluated on both synthetic and real data.
Tasks
Published	2019-07-24
URL	https://arxiv.org/abs/1907.10557v2
PDF	https://arxiv.org/pdf/1907.10557v2.pdf
PWC	https://paperswithcode.com/paper/maximum-likelihood-estimation-for-disk-image
Repo
Framework

Memory-efficient Learning for Large-scale Computational Imaging – NeurIPS deep inverse workshop


Title	Memory-efficient Learning for Large-scale Computational Imaging – NeurIPS deep inverse workshop
Authors	Michael Kellman, Jon Tamir, Emrah Boston, Michael Lustig, Laura Waller
Abstract	Computational imaging systems jointly design computation and hardware to retrieve information which is not traditionally accessible with standard imaging systems. Recently, critical aspects such as experimental design and image priors are optimized through deep neural networks formed by the unrolled iterations of classical physics-based reconstructions (termed physics-based networks). However, for real-world large-scale systems, computing gradients via backpropagation restricts learning due to memory limitations of graphical processing units. In this work, we propose a memory-efficient learning procedure that exploits the reversibility of the network’s layers to enable data-driven design for large-scale computational imaging. We demonstrate our methods practicality on two large-scale systems: super-resolution optical microscopy and multi-channel magnetic resonance imaging.
Tasks	Super-Resolution
Published	2019-12-11
URL	https://arxiv.org/abs/1912.05098v2
PDF	https://arxiv.org/pdf/1912.05098v2.pdf
PWC	https://paperswithcode.com/paper/memory-efficient-learning-for-large-scale
Repo
Framework

Learning Policies through Quantile Regression


Title	Learning Policies through Quantile Regression
Authors	Oliver Richter, Roger Wattenhofer
Abstract	Policy gradient based reinforcement learning algorithms coupled with neural networks have shown success in learning complex policies in the model free continuous action space control setting. However, explicitly parameterized policies are limited by the scope of the chosen parametric probability distribution. We show that alternatively to the likelihood based policy gradient, a related objective can be optimized through advantage weighted quantile regression. Our approach models the policy implicitly in the network, which gives the agent the freedom to approximate any distribution in each action dimension, not limiting its capabilities to the commonly used unimodal Gaussian parameterization. This broader spectrum of policies makes our algorithm suitable for problems where Gaussian policies cannot fit the optimal policy. Moreover, our results on the MuJoCo physics simulator benchmarks are comparable or superior to state-of-the-art on-policy methods.
Tasks
Published	2019-06-27
URL	https://arxiv.org/abs/1906.11941v2
PDF	https://arxiv.org/pdf/1906.11941v2.pdf
PWC	https://paperswithcode.com/paper/quantile-regression-deep-reinforcement
Repo
Framework

Reduced Focal Loss: 1st Place Solution to xView object detection in Satellite Imagery


Title	Reduced Focal Loss: 1st Place Solution to xView object detection in Satellite Imagery
Authors	Nikolay Sergievskiy, Alexander Ponamarev
Abstract	This paper describes our approach to the DIUx xView 2018 Detection Challenge [1]. This challenge focuses on a new satellite imagery dataset. The dataset contains 60 object classes that are highly imbalanced. Due to the imbalanced nature of the dataset, the training process becomes significantly more challenging. To address this problem, we introduce a novel Reduced Focal Loss function, which brought us 1st place in the DIUx xView 2018 Detection Challenge.
Tasks	Object Detection
Published	2019-03-04
URL	http://arxiv.org/abs/1903.01347v2
PDF	http://arxiv.org/pdf/1903.01347v2.pdf
PWC	https://paperswithcode.com/paper/reduced-focal-loss-1st-place-solution-to
Repo
Framework

Trait of Gait: A Survey on Gait Biometrics


Title	Trait of Gait: A Survey on Gait Biometrics
Authors	Ebenezer R. H. P. Isaac, Susan Elias, Srinivasan Rajagopalan, K. S. Easwarakumar
Abstract	Gait analysis is the study of the systematic methods that assess and quantify animal locomotion. The research on gait analysis has considerably evolved through time. It was an ancient art, and it still finds its application today in modern science and medicine. This paper describes how one’s gait can be used as a biometric. It shall diversely cover salient research done within the field and explain the nuances and advances in each type of gait analysis. The prominent methods of gait recognition from the early era to the state of the art are covered. This survey also reviews the various gait datasets. The overall aim of this study is to provide a concise roadmap for anyone who wishes to do research in the field of gait biometrics.
Tasks	Gait Recognition
Published	2019-03-26
URL	http://arxiv.org/abs/1903.10744v1
PDF	http://arxiv.org/pdf/1903.10744v1.pdf
PWC	https://paperswithcode.com/paper/trait-of-gait-a-survey-on-gait-biometrics
Repo
Framework

Kite: Automatic speech recognition for unmanned aerial vehicles


Title	Kite: Automatic speech recognition for unmanned aerial vehicles
Authors	Dan Oneata, Horia Cucu
Abstract	This paper addresses the problem of building a speech recognition system attuned to the control of unmanned aerial vehicles (UAVs). Even though UAVs are becoming widespread, the task of creating voice interfaces for them is largely unaddressed. To this end, we introduce a multi-modal evaluation dataset for UAV control, consisting of spoken commands and associated images, which represent the visual context of what the UAV “sees” when the pilot utters the command. We provide baseline results and address two research directions: (i) how robust the language models are, given an incomplete list of commands at train time; (ii) how to incorporate visual information in the language model. We find that recurrent neural networks (RNNs) are a solution to both tasks: they can be successfully adapted using a small number of commands and they can be extended to use visual cues. Our results show that the image-based RNN outperforms its text-only counterpart even if the command-image training associations are automatically generated and inherently imperfect. The dataset and our code are available at http://kite.speed.pub.ro.
Tasks	Language Modelling, Speech Recognition
Published	2019-07-02
URL	https://arxiv.org/abs/1907.01195v1
PDF	https://arxiv.org/pdf/1907.01195v1.pdf
PWC	https://paperswithcode.com/paper/kite-automatic-speech-recognition-for
Repo
Framework

AMP0: Species-Specific Prediction of Anti-microbial Peptides using Zero and Few Shot Learning


Title	AMP0: Species-Specific Prediction of Anti-microbial Peptides using Zero and Few Shot Learning
Authors	Sadaf Gull, Fayyaz Minhas
Abstract	The evolution of drug-resistant microbial species is one of the major challenges to global health. The development of new antimicrobial treatments such as antimicrobial peptides needs to be accelerated to combat this threat. However, the discovery of novel antimicrobial peptides is hampered by low-throughput biochemical assays. Computational techniques can be used for rapid screening of promising antimicrobial peptide candidates prior to testing in the wet lab. The vast majority of existing antimicrobial peptide predictors are non-targeted in nature, i.e., they can predict whether a given peptide sequence is antimicrobial, but they are unable to predict whether the sequence can target a particular microbial species. In this work, we have developed a targeted antimicrobial peptide activity predictor that can predict whether a peptide is effective against a given microbial species or not. This has been made possible through zero-shot and few-shot machine learning. The proposed predictor called AMP0 takes in the peptide amino acid sequence and any N/C-termini modifications together with the genomic sequence of a target microbial species to generate targeted predictions. It is important to note that the proposed method can generate predictions for species that are not part of its training set. The accuracy of predictions for novel test species can be further improved by providing a few example peptides for that species. Our computational cross-validation results show that the pro-posed scheme is particularly effective for targeted antimicrobial prediction in comparison to existing approaches and can be used for screening potential antimicrobial peptides in a targeted manner especially for cases in which the number of training examples is small. The webserver of the method is available at http://ampzero.pythonanywhere.com.
Tasks	Few-Shot Learning
Published	2019-10-28
URL	https://arxiv.org/abs/1911.06106v1
PDF	https://arxiv.org/pdf/1911.06106v1.pdf
PWC	https://paperswithcode.com/paper/amp0-species-specific-prediction-of-anti
Repo
Framework