May 7, 2019

2884 words 14 mins read

Paper Group ANR 153

Revealing the Hidden Patterns of News Photos: Analysis of Millions of News Photos Using GDELT and Deep Learning-based Vision APIs. Sample Complexity of Automated Mechanism Design. On the performance of different mutation operators of a subpopulation-based genetic algorithm for multi-robot task allocation problems. Polynomial-time Tensor Decompositi …

Revealing the Hidden Patterns of News Photos: Analysis of Millions of News Photos Using GDELT and Deep Learning-based Vision APIs


Title	Revealing the Hidden Patterns of News Photos: Analysis of Millions of News Photos Using GDELT and Deep Learning-based Vision APIs
Authors	Haewoon Kwak, Jisun An
Abstract	In this work, we analyze more than two million news photos published in January 2016. We demonstrate i) which objects appear the most in news photos; ii) what the sentiments of news photos are; iii) whether the sentiment of news photos is aligned with the tone of the text; iv) how gender is treated; and v) how differently political candidates are portrayed. To our best knowledge, this is the first large-scale study of news photo contents using deep learning-based vision APIs.
Tasks
Published	2016-03-15
URL	http://arxiv.org/abs/1603.04531v2
PDF	http://arxiv.org/pdf/1603.04531v2.pdf
PWC	https://paperswithcode.com/paper/revealing-the-hidden-patterns-of-news-photos
Repo
Framework

Sample Complexity of Automated Mechanism Design


Title	Sample Complexity of Automated Mechanism Design
Authors	Maria-Florina Balcan, Tuomas Sandholm, Ellen Vitercik
Abstract	The design of revenue-maximizing combinatorial auctions, i.e. multi-item auctions over bundles of goods, is one of the most fundamental problems in computational economics, unsolved even for two bidders and two items for sale. In the traditional economic models, it is assumed that the bidders’ valuations are drawn from an underlying distribution and that the auction designer has perfect knowledge of this distribution. Despite this strong and oftentimes unrealistic assumption, it is remarkable that the revenue-maximizing combinatorial auction remains unknown. In recent years, automated mechanism design has emerged as one of the most practical and promising approaches to designing high-revenue combinatorial auctions. The most scalable automated mechanism design algorithms take as input samples from the bidders’ valuation distribution and then search for a high-revenue auction in a rich auction class. In this work, we provide the first sample complexity analysis for the standard hierarchy of deterministic combinatorial auction classes used in automated mechanism design. In particular, we provide tight sample complexity bounds on the number of samples needed to guarantee that the empirical revenue of the designed mechanism on the samples is close to its expected revenue on the underlying, unknown distribution over bidder valuations, for each of the auction classes in the hierarchy. In addition to helping set automated mechanism design on firm foundations, our results also push the boundaries of learning theory. In particular, the hypothesis functions used in our contexts are defined through multi-stage combinatorial optimization procedures, rather than simple decision boundaries, as are common in machine learning.
Tasks	Combinatorial Optimization
Published	2016-06-13
URL	http://arxiv.org/abs/1606.04145v1
PDF	http://arxiv.org/pdf/1606.04145v1.pdf
PWC	https://paperswithcode.com/paper/sample-complexity-of-automated-mechanism
Repo
Framework

On the performance of different mutation operators of a subpopulation-based genetic algorithm for multi-robot task allocation problems


Title	On the performance of different mutation operators of a subpopulation-based genetic algorithm for multi-robot task allocation problems
Authors	Chun Liu, Andreas Kroll
Abstract	The performance of different mutation operators is usually evaluated in conjunc-tion with specific parameter settings of genetic algorithms and target problems. Most studies focus on the classical genetic algorithm with different parameters or on solving unconstrained combinatorial optimization problems such as the traveling salesman problems. In this paper, a subpopulation-based genetic al-gorithm that uses only mutation and selection is developed to solve multi-robot task allocation problems. The target problems are constrained combinatorial optimization problems, and are more complex if cooperative tasks are involved as these introduce additional spatial and temporal constraints. The proposed genetic algorithm can obtain better solutions than classical genetic algorithms with tournament selection and partially mapped crossover. The performance of different mutation operators in solving problems without/with cooperative tasks is evaluated. The results imply that inversion mutation performs better than others when solving problems without cooperative tasks, and the swap-inversion combination performs better than others when solving problems with cooperative tasks.
Tasks	Combinatorial Optimization
Published	2016-06-02
URL	http://arxiv.org/abs/1606.00601v1
PDF	http://arxiv.org/pdf/1606.00601v1.pdf
PWC	https://paperswithcode.com/paper/on-the-performance-of-different-mutation
Repo
Framework

Polynomial-time Tensor Decompositions with Sum-of-Squares


Title	Polynomial-time Tensor Decompositions with Sum-of-Squares
Authors	Tengyu Ma, Jonathan Shi, David Steurer
Abstract	We give new algorithms based on the sum-of-squares method for tensor decomposition. Our results improve the best known running times from quasi-polynomial to polynomial for several problems, including decomposing random overcomplete 3-tensors and learning overcomplete dictionaries with constant relative sparsity. We also give the first robust analysis for decomposing overcomplete 4-tensors in the smoothed analysis model. A key ingredient of our analysis is to establish small spectral gaps in moment matrices derived from solutions to sum-of-squares relaxations. To enable this analysis we augment sum-of-squares relaxations with spectral analogs of maximum entropy constraints.
Tasks
Published	2016-10-06
URL	http://arxiv.org/abs/1610.01980v1
PDF	http://arxiv.org/pdf/1610.01980v1.pdf
PWC	https://paperswithcode.com/paper/polynomial-time-tensor-decompositions-with
Repo
Framework

Local Perturb-and-MAP for Structured Prediction


Title	Local Perturb-and-MAP for Structured Prediction
Authors	Gedas Bertasius, Qiang Liu, Lorenzo Torresani, Jianbo Shi
Abstract	Conditional random fields (CRFs) provide a powerful tool for structured prediction, but cast significant challenges in both the learning and inference steps. Approximation techniques are widely used in both steps, which should be considered jointly to guarantee good performance (a.k.a. “inferning”). Perturb-and-MAP models provide a promising alternative to CRFs, but require global combinatorial optimization and hence they are usable only on specific models. In this work, we present a new Local Perturb-and-MAP (locPMAP) framework that replaces the global optimization with a local optimization by exploiting our observed connection between locPMAP and the pseudolikelihood of the original CRF model. We test our approach on three different vision tasks and show that our method achieves consistently improved performance over other approximate inference techniques optimized to a pseudolikelihood objective. Additionally, we demonstrate that we can integrate our method in the fully convolutional network framework to increase our model’s complexity. Finally, our observed connection between locPMAP and the pseudolikelihood leads to a novel perspective for understanding and using pseudolikelihood.
Tasks	Combinatorial Optimization, Structured Prediction
Published	2016-05-24
URL	http://arxiv.org/abs/1605.07686v2
PDF	http://arxiv.org/pdf/1605.07686v2.pdf
PWC	https://paperswithcode.com/paper/local-perturb-and-map-for-structured
Repo
Framework

Spectral decomposition method of dialog state tracking via collective matrix factorization


Title	Spectral decomposition method of dialog state tracking via collective matrix factorization
Authors	Julien Perez
Abstract	The task of dialog management is commonly decomposed into two sequential subtasks: dialog state tracking and dialog policy learning. In an end-to-end dialog system, the aim of dialog state tracking is to accurately estimate the true dialog state from noisy observations produced by the speech recognition and the natural language understanding modules. The state tracking task is primarily meant to support a dialog policy. From a probabilistic perspective, this is achieved by maintaining a posterior distribution over hidden dialog states composed of a set of context dependent variables. Once a dialog policy is learned, it strives to select an optimal dialog act given the estimated dialog state and a defined reward function. This paper introduces a novel method of dialog state tracking based on a bilinear algebric decomposition model that provides an efficient inference schema through collective matrix factorization. We evaluate the proposed approach on the second Dialog State Tracking Challenge (DSTC-2) dataset and we show that the proposed tracker gives encouraging results compared to the state-of-the-art trackers that participated in this standard benchmark. Finally, we show that the prediction schema is computationally efficient in comparison to the previous approaches.
Tasks	Speech Recognition
Published	2016-06-16
URL	http://arxiv.org/abs/1606.05286v1
PDF	http://arxiv.org/pdf/1606.05286v1.pdf
PWC	https://paperswithcode.com/paper/spectral-decomposition-method-of-dialog-state
Repo
Framework

Video Processing from Electro-optical Sensors for Object Detection and Tracking in Maritime Environment: A Survey


Title	Video Processing from Electro-optical Sensors for Object Detection and Tracking in Maritime Environment: A Survey
Authors	D. K. Prasad, D. Rajan, L. Rachmawati, E. Rajabaly, C. Quek
Abstract	We present a survey on maritime object detection and tracking approaches, which are essential for the development of a navigational system for autonomous ships. The electro-optical (EO) sensor considered here is a video camera that operates in the visible or the infrared spectra, which conventionally complement radar and sonar and have demonstrated effectiveness for situational awareness at sea has demonstrated its effectiveness over the last few years. This paper provides a comprehensive overview of various approaches of video processing for object detection and tracking in the maritime environment. We follow an approach-based taxonomy wherein the advantages and limitations of each approach are compared. The object detection system consists of the following modules: horizon detection, static background subtraction and foreground segmentation. Each of these has been studied extensively in maritime situations and has been shown to be challenging due to the presence of background motion especially due to waves and wakes. The main processes involved in object tracking include video frame registration, dynamic background subtraction, and the object tracking algorithm itself. The challenges for robust tracking arise due to camera motion, dynamic background and low contrast of tracked object, possibly due to environmental degradation. The survey also discusses multisensor approaches and commercial maritime systems that use EO sensors. The survey also highlights methods from computer vision research which hold promise to perform well in maritime EO data processing. Performance of several maritime and computer vision techniques is evaluated on newly proposed Singapore Maritime Dataset.
Tasks	Object Detection, Object Tracking
Published	2016-11-17
URL	http://arxiv.org/abs/1611.05842v1
PDF	http://arxiv.org/pdf/1611.05842v1.pdf
PWC	https://paperswithcode.com/paper/video-processing-from-electro-optical-sensors
Repo
Framework

Accelerating Deep Convolutional Networks using low-precision and sparsity


Title	Accelerating Deep Convolutional Networks using low-precision and sparsity
Authors	Ganesh Venkatesh, Eriko Nurvitadhi, Debbie Marr
Abstract	We explore techniques to significantly improve the compute efficiency and performance of Deep Convolution Networks without impacting their accuracy. To improve the compute efficiency, we focus on achieving high accuracy with extremely low-precision (2-bit) weight networks, and to accelerate the execution time, we aggressively skip operations on zero-values. We achieve the highest reported accuracy of 76.6% Top-1/93% Top-5 on the Imagenet object classification challenge with low-precision network\footnote{github release of the source code coming soon} while reducing the compute requirement by ~3x compared to a full-precision network that achieves similar accuracy. Furthermore, to fully exploit the benefits of our low-precision networks, we build a deep learning accelerator core, dLAC, that can achieve up to 1 TFLOP/mm^2 equivalent for single-precision floating-point operations (~2 TFLOP/mm^2 for half-precision).
Tasks	Object Classification
Published	2016-10-02
URL	http://arxiv.org/abs/1610.00324v1
PDF	http://arxiv.org/pdf/1610.00324v1.pdf
PWC	https://paperswithcode.com/paper/accelerating-deep-convolutional-networks
Repo
Framework

Geometry in Active Learning for Binary and Multi-class Image Segmentation


Title	Geometry in Active Learning for Binary and Multi-class Image Segmentation
Authors	Ksenia Konyushkova, Raphael Sznitman, Pascal Fua
Abstract	We propose an active learning approach to image segmentation that exploits geometric priors to speed up and streamline the annotation process. It can be applied for both background-foreground and multi-class segmentation tasks in 2D images and 3D image volumes. Our approach combines geometric smoothness priors in the image space with more traditional uncertainty measures to estimate which pixels or voxels are the most informative, and thus should to be annotated next. For multi-class settings, we additionally introduce two novel criteria for uncertainty. In the 3D case, we use the resulting uncertainty measure to select voxels lying on a planar patch, which makes batch annotation much more convenient for the end user compared to the setting where voxels are randomly distributed in a volume. The planar patch is found using a branch-and-bound algorithm that looks for a 2D patch in a 3D volume where the most informative instances are located. We evaluate our approach on Electron Microscopy and Magnetic Resonance image volumes, as well as on regular images of horses and faces. We demonstrate a substantial performance increase over other approaches thanks to the use of geometric priors.
Tasks	Active Learning, Semantic Segmentation
Published	2016-06-29
URL	http://arxiv.org/abs/1606.09029v4
PDF	http://arxiv.org/pdf/1606.09029v4.pdf
PWC	https://paperswithcode.com/paper/geometry-in-active-learning-for-binary-and
Repo
Framework

Automated 5-year Mortality Prediction using Deep Learning and Radiomics Features from Chest Computed Tomography


Title	Automated 5-year Mortality Prediction using Deep Learning and Radiomics Features from Chest Computed Tomography
Authors	Gustavo Carneiro, Luke Oakden-Rayner, Andrew P. Bradley, Jacinto Nascimento, Lyle Palmer
Abstract	We propose new methods for the prediction of 5-year mortality in elderly individuals using chest computed tomography (CT). The methods consist of a classifier that performs this prediction using a set of features extracted from the CT image and segmentation maps of multiple anatomic structures. We explore two approaches: 1) a unified framework based on deep learning, where features and classifier are automatically learned in a single optimisation process; and 2) a multi-stage framework based on the design and selection/extraction of hand-crafted radiomics features, followed by the classifier learning process. Experimental results, based on a dataset of 48 annotated chest CTs, show that the deep learning model produces a mean 5-year mortality prediction accuracy of 68.5%, while radiomics produces a mean accuracy that varies between 56% to 66% (depending on the feature selection/extraction method and classifier). The successful development of the proposed models has the potential to make a profound impact in preventive and personalised healthcare.
Tasks	Computed Tomography (CT), Feature Selection, Mortality Prediction
Published	2016-07-01
URL	http://arxiv.org/abs/1607.00267v1
PDF	http://arxiv.org/pdf/1607.00267v1.pdf
PWC	https://paperswithcode.com/paper/automated-5-year-mortality-prediction-using
Repo
Framework

Face Recognition: Perspectives from the Real-World


Title	Face Recognition: Perspectives from the Real-World
Authors	Bappaditya Mandal
Abstract	In this paper, we analyze some of our real-world deployment of face recognition (FR) systems for various applications and discuss the gaps between expectations of the user and what the system can deliver. We evaluate some of our proposed algorithms with ad-hoc modifications for applications such as FR on wearable devices (like Google Glass), monitoring of elderly people in senior citizens centers, FR of children in child care centers and face matching between a scanned IC/passport face image and a few live webcam images for automatic hotel/resort checkouts. We describe each of these applications, the challenges involved and proposed solutions. Since FR is intuitive in nature and we human beings use it for interactions with the outside world, people have high expectations of its performance in real-world scenarios. However, we analyze and discuss here that it is not the case, machine recognition of faces for each of these applications poses unique challenges and demands specific research components so as to adapt in the actual sites.
Tasks	Face Recognition
Published	2016-02-09
URL	http://arxiv.org/abs/1602.02999v1
PDF	http://arxiv.org/pdf/1602.02999v1.pdf
PWC	https://paperswithcode.com/paper/face-recognition-perspectives-from-the-real
Repo
Framework

A New Training Method for Feedforward Neural Networks Based on Geometric Contraction Property of Activation Functions


Title	A New Training Method for Feedforward Neural Networks Based on Geometric Contraction Property of Activation Functions
Authors	Petre Birtea, Cosmin Cernazanu-Glavan, Alexandru Sisu
Abstract	We propose a new training method for a feedforward neural network having the activation functions with the geometric contraction property. The method consists of constructing a new functional that is less nonlinear in comparison with the classical functional by removing the nonlinearity of the activation function from the output layer. We validate this new method by a series of experiments that show an improved learning speed and better classification error.
Tasks
Published	2016-06-20
URL	http://arxiv.org/abs/1606.05990v2
PDF	http://arxiv.org/pdf/1606.05990v2.pdf
PWC	https://paperswithcode.com/paper/a-new-training-method-for-feedforward-neural
Repo
Framework

Bayesian Neighbourhood Component Analysis


Title	Bayesian Neighbourhood Component Analysis
Authors	Dong Wang, Xiaoyang Tan
Abstract	Learning a good distance metric in feature space potentially improves the performance of the KNN classifier and is useful in many real-world applications. Many metric learning algorithms are however based on the point estimation of a quadratic optimization problem, which is time-consuming, susceptible to overfitting, and lack a natural mechanism to reason with parameter uncertainty, an important property useful especially when the training set is small and/or noisy. To deal with these issues, we present a novel Bayesian metric learning method, called Bayesian NCA, based on the well-known Neighbourhood Component Analysis method, in which the metric posterior is characterized by the local label consistency constraints of observations, encoded with a similarity graph instead of independent pairwise constraints. For efficient Bayesian optimization, we explore the variational lower bound over the log-likelihood of the original NCA objective. Experiments on several publicly available datasets demonstrate that the proposed method is able to learn robust metric measures from small size dataset and/or from challenging training set with labels contaminated by errors. The proposed method is also shown to outperform a previous pairwise constrained Bayesian metric learning method.
Tasks	Metric Learning
Published	2016-04-08
URL	http://arxiv.org/abs/1604.02354v1
PDF	http://arxiv.org/pdf/1604.02354v1.pdf
PWC	https://paperswithcode.com/paper/bayesian-neighbourhood-component-analysis
Repo
Framework

QPass: a Merit-based Evaluation of Soccer Passes


Title	QPass: a Merit-based Evaluation of Soccer Passes
Authors	Laszlo Gyarmati, Rade Stanojevic
Abstract	Quantitative analysis of soccer players’ passing ability focuses on descriptive statistics without considering the players’ real contribution to the passing and ball possession strategy of their team. Which player is able to help the build-up of an attack, or to maintain the possession of the ball? We introduce a novel methodology called QPass to answer questions like these quantitatively. Based on the analysis of an entire season, we rank the players based on the intrinsic value of their passes using QPass. We derive an album of pass trajectories for different gaming styles. Our methodology reveals a quite counterintuitive paradigm: losing the ball possession could lead to better chances to win a game.
Tasks
Published	2016-08-08
URL	http://arxiv.org/abs/1608.03532v1
PDF	http://arxiv.org/pdf/1608.03532v1.pdf
PWC	https://paperswithcode.com/paper/qpass-a-merit-based-evaluation-of-soccer
Repo
Framework

Ashwin: Plug-and-Play System for Machine-Human Image Annotation


Title	Ashwin: Plug-and-Play System for Machine-Human Image Annotation
Authors	Anand Sriraman, Mandar Kulkarni, Rahul Kumar, Kanika Kalra, Purushotam Radadia, Shirish Karande
Abstract	We present an end-to-end machine-human image annotation system where each component can be attached in a plug-and-play fashion. These components include Feature Extraction, Machine Classifier, Task Sampling and Crowd Consensus.
Tasks
Published	2016-09-08
URL	http://arxiv.org/abs/1609.02271v2
PDF	http://arxiv.org/pdf/1609.02271v2.pdf
PWC	https://paperswithcode.com/paper/ashwin-plug-and-play-system-for-machine-human
Repo
Framework