May 6, 2019

2669 words 13 mins read

Paper Group ANR 260

Rotation Invariant Angular Descriptor Via A Bandlimited Gaussian-like Kernel. Understanding and Mapping Natural Beauty. Application of multiview techniques to NHANES dataset. Semi-Supervised Classification Based on Classification from Positive and Unlabeled Data. Automated Prediction of Temporal Relations. Super-Resolution Reconstruction of Electri …

Rotation Invariant Angular Descriptor Via A Bandlimited Gaussian-like Kernel


Title	Rotation Invariant Angular Descriptor Via A Bandlimited Gaussian-like Kernel
Authors	Michael T. McCann, Matthew Fickus, Jelena Kovacevic
Abstract	We present a new smooth, Gaussian-like kernel that allows the kernel density estimate for an angular distribution to be exactly represented by a finite number of its Fourier series coefficients. Distributions of angular quantities, such as gradients, are a central part of several state-of-the-art image processing algorithms, but these distributions are usually described via histograms and therefore lack rotation invariance due to binning artifacts. Replacing histograming with kernel density estimation removes these binning artifacts and can provide a finite-dimensional descriptor of the distribution, provided that the kernel is selected to be bandlimited. In this paper, we present a new band-limited kernel that has the added advantage of being Gaussian-like in the angular domain. We then show that it compares favorably to gradient histograms for patch matching, person detection, and texture segmentation.
Tasks	Density Estimation, Human Detection
Published	2016-06-08
URL	http://arxiv.org/abs/1606.02753v1
PDF	http://arxiv.org/pdf/1606.02753v1.pdf
PWC	https://paperswithcode.com/paper/rotation-invariant-angular-descriptor-via-a
Repo
Framework

Understanding and Mapping Natural Beauty


Title	Understanding and Mapping Natural Beauty
Authors	Scott Workman, Richard Souvenir, Nathan Jacobs
Abstract	While natural beauty is often considered a subjective property of images, in this paper, we take an objective approach and provide methods for quantifying and predicting the scenicness of an image. Using a dataset containing hundreds of thousands of outdoor images captured throughout Great Britain with crowdsourced ratings of natural beauty, we propose an approach to predict scenicness which explicitly accounts for the variance of human ratings. We demonstrate that quantitative measures of scenicness can benefit semantic image understanding, content-aware image processing, and a novel application of cross-view mapping, where the sparsity of ground-level images can be addressed by incorporating unlabeled overhead images in the training and prediction steps. For each application, our methods for scenicness prediction result in quantitative and qualitative improvements over baseline approaches.
Tasks
Published	2016-12-09
URL	http://arxiv.org/abs/1612.03142v2
PDF	http://arxiv.org/pdf/1612.03142v2.pdf
PWC	https://paperswithcode.com/paper/understanding-and-mapping-natural-beauty
Repo
Framework

Application of multiview techniques to NHANES dataset


Title	Application of multiview techniques to NHANES dataset
Authors	Aileme Omogbai
Abstract	Disease prediction or classification using health datasets involve using well-known predictors associated with the disease as features for the models. This study considers multiple data components of an individual’s health, using the relationship between variables to generate features that may improve the performance of disease classification models. In order to capture information from different aspects of the data, this project uses a multiview learning approach, using Canonical Correlation Analysis (CCA), a technique that finds projections with maximum correlations between two data views. Data categories collected from the NHANES survey (1999-2014) are used as views to learn the multiview representations. The usefulness of the representations is demonstrated by applying them as features in a Diabetes classification task.
Tasks	Disease Prediction, Multiview Learning
Published	2016-08-16
URL	http://arxiv.org/abs/1608.04783v1
PDF	http://arxiv.org/pdf/1608.04783v1.pdf
PWC	https://paperswithcode.com/paper/application-of-multiview-techniques-to-nhanes
Repo
Framework

Semi-Supervised Classification Based on Classification from Positive and Unlabeled Data


Title	Semi-Supervised Classification Based on Classification from Positive and Unlabeled Data
Authors	Tomoya Sakai, Marthinus Christoffel du Plessis, Gang Niu, Masashi Sugiyama
Abstract	Most of the semi-supervised classification methods developed so far use unlabeled data for regularization purposes under particular distributional assumptions such as the cluster assumption. In contrast, recently developed methods of classification from positive and unlabeled data (PU classification) use unlabeled data for risk evaluation, i.e., label information is directly extracted from unlabeled data. In this paper, we extend PU classification to also incorporate negative data and propose a novel semi-supervised classification approach. We establish generalization error bounds for our novel methods and show that the bounds decrease with respect to the number of unlabeled data without the distributional assumptions that are required in existing semi-supervised classification methods. Through experiments, we demonstrate the usefulness of the proposed methods.
Tasks
Published	2016-05-23
URL	http://arxiv.org/abs/1605.06955v4
PDF	http://arxiv.org/pdf/1605.06955v4.pdf
PWC	https://paperswithcode.com/paper/semi-supervised-classification-based-on
Repo
Framework

Automated Prediction of Temporal Relations


Title	Automated Prediction of Temporal Relations
Authors	Amol S Patwardhan, Jacob Badeaux, Siavash, Gerald M Knapp
Abstract	Background: There has been growing research interest in automated answering of questions or generation of summary of free form text such as news article. In order to implement this task, the computer should be able to identify the sequence of events, duration of events, time at which event occurred and the relationship type between event pairs, time pairs or event-time pairs. Specific Problem: It is important to accurately identify the relationship type between combinations of event and time before the temporal ordering of events can be defined. The machine learning approach taken in Mani et. al (2006) provides an accuracy of only 62.5 on the baseline data from TimeBank. The researchers used maximum entropy classifier in their methodology. TimeML uses the TLINK annotation to tag a relationship type between events and time. The time complexity is quadratic when it comes to tagging documents with TLINK using human annotation. This research proposes using decision tree and parsing to improve the relationship type tagging. This research attempts to solve the gaps in human annotation by automating the task of relationship type tagging in an attempt to improve the accuracy of event and time relationship in annotated documents. Scope information: The documents from the domain of news will be used. The tagging will be performed within the same document and not across documents. The relationship types will be identified only for a pair of event and time and not a chain of events. The research focuses on documents tagged using the TimeML specification which contains tags such as EVENT, TLINK, and TIMEX. Each tag has attributes such as identifier, relation, POS, time etc.
Tasks
Published	2016-07-22
URL	http://arxiv.org/abs/1607.06560v1
PDF	http://arxiv.org/pdf/1607.06560v1.pdf
PWC	https://paperswithcode.com/paper/automated-prediction-of-temporal-relations
Repo
Framework

Super-Resolution Reconstruction of Electrical Impedance Tomography Images


Title	Super-Resolution Reconstruction of Electrical Impedance Tomography Images
Authors	Ricardo A. Borsoi, Julio C. C. Aya, Guilherme H. Costa, José C. M. Bermudez
Abstract	Electrical Impedance Tomography (EIT) systems are becoming popular because they present several advantages over competing systems. However, EIT leads to images with very low resolution. Moreover, the nonuniform sampling characteristic of EIT precludes the straightforward application of traditional image ruper-resolution techniques. In this work, we propose a resampling based Super-Resolution method for EIT image quality improvement. Preliminary results show that the proposed technique can lead to substantial improvements in EIT image resolution, making it more competitive with other technologies.
Tasks	Super-Resolution
Published	2016-12-30
URL	http://arxiv.org/abs/1701.00031v3
PDF	http://arxiv.org/pdf/1701.00031v3.pdf
PWC	https://paperswithcode.com/paper/super-resolution-reconstruction-of-electrical
Repo
Framework

Generation of discrete random variables in scalable frameworks


Title	Generation of discrete random variables in scalable frameworks
Authors	Giacomo Aletti
Abstract	In this paper, we face the problem of simulating discrete random variables with general and varying distributions in a scalable framework, where fully parallelizable operations should be preferred. The new paradigm is inspired by the context of discrete choice models. Compared to classical algorithms, we add parallelized randomness, and we leave the final simulation of the random variable to a single associative operation. We characterize the set of algorithms that work in this way, and those algorithms that may have an additive or multiplicative local noise. As a consequence, we could define a natural way to solve some popular simulation problems.
Tasks
Published	2016-11-21
URL	http://arxiv.org/abs/1611.07103v3
PDF	http://arxiv.org/pdf/1611.07103v3.pdf
PWC	https://paperswithcode.com/paper/generation-of-discrete-random-variables-in
Repo
Framework

Poincaré inequalities on intervals – application to sensitivity analysis


Title	Poincaré inequalities on intervals – application to sensitivity analysis
Authors	Olivier Roustant, Franck Barthe, Bertrand Iooss
Abstract	The development of global sensitivity analysis of numerical model outputs has recently raised new issues on 1-dimensional Poincar'e inequalities. Typically two kind of sensitivity indices are linked by a Poincar'e type inequality, which provide upper bounds of the most interpretable index by using the other one, cheaper to compute. This allows performing a low-cost screening of unessential variables. The efficiency of this screening then highly depends on the accuracy of the upper bounds in Poincar'e inequalities. The novelty in the questions concern the wide range of probability distributions involved, which are often truncated on intervals. After providing an overview of the existing knowledge and techniques, we add some theory about Poincar'e constants on intervals, with improvements for symmetric intervals. Then we exploit the spectral interpretation for computing exact value of Poincar'e constants of any admissible distribution on a given interval. We give semi-analytical results for some frequent distributions (truncated exponential, triangular, truncated normal), and present a numerical method in the general case. Finally, an application is made to a hydrological problem, showing the benefits of the new results in Poincar'e inequalities to sensitivity analysis.
Tasks
Published	2016-12-12
URL	http://arxiv.org/abs/1612.03689v1
PDF	http://arxiv.org/pdf/1612.03689v1.pdf
PWC	https://paperswithcode.com/paper/poincare-inequalities-on-intervals
Repo
Framework

Neural Network Based Next-Song Recommendation


Title	Neural Network Based Next-Song Recommendation
Authors	Kai-Chun Hsu, Szu-Yu Chou, Yi-Hsuan Yang, Tai-Shih Chi
Abstract	Recently, the next-item/basket recommendation system, which considers the sequential relation between bought items, has drawn attention of researchers. The utilization of sequential patterns has boosted performance on several kinds of recommendation tasks. Inspired by natural language processing (NLP) techniques, we propose a novel neural network (NN) based next-song recommender, CNN-rec, in this paper. Then, we compare the proposed system with several NN based and classic recommendation systems on the next-song recommendation task. Verification results indicate the proposed system outperforms classic systems and has comparable performance with the state-of-the-art system.
Tasks	Recommendation Systems
Published	2016-06-24
URL	http://arxiv.org/abs/1606.07722v1
PDF	http://arxiv.org/pdf/1606.07722v1.pdf
PWC	https://paperswithcode.com/paper/neural-network-based-next-song-recommendation
Repo
Framework

Left/Right Hand Segmentation in Egocentric Videos


Title	Left/Right Hand Segmentation in Egocentric Videos
Authors	Alejandro Betancourt, Pietro Morerio, Emilia Barakova, Lucio Marcenaro, Matthias Rauterberg, Carlo Regazzoni
Abstract	Wearable cameras allow people to record their daily activities from a user-centered (First Person Vision) perspective. Due to their favorable location, wearable cameras frequently capture the hands of the user, and may thus represent a promising user-machine interaction tool for different applications. Existent First Person Vision methods handle hand segmentation as a background-foreground problem, ignoring two important facts: i) hands are not a single “skin-like” moving element, but a pair of interacting cooperative entities, ii) close hand interactions may lead to hand-to-hand occlusions and, as a consequence, create a single hand-like segment. These facts complicate a proper understanding of hand movements and interactions. Our approach extends traditional background-foreground strategies, by including a hand-identification step (left-right) based on a Maxwell distribution of angle and position. Hand-to-hand occlusions are addressed by exploiting temporal superpixels. The experimental results show that, in addition to a reliable left/right hand-segmentation, our approach considerably improves the traditional background-foreground hand-segmentation.
Tasks	Hand Segmentation
Published	2016-07-21
URL	http://arxiv.org/abs/1607.06264v1
PDF	http://arxiv.org/pdf/1607.06264v1.pdf
PWC	https://paperswithcode.com/paper/leftright-hand-segmentation-in-egocentric
Repo
Framework

AudioPairBank: Towards A Large-Scale Tag-Pair-Based Audio Content Analysis


Title	AudioPairBank: Towards A Large-Scale Tag-Pair-Based Audio Content Analysis
Authors	Sebastian Sager, Benjamin Elizalde, Damian Borth, Christian Schulze, Bhiksha Raj, Ian Lane
Abstract	Recently, sound recognition has been used to identify sounds, such as car and river. However, sounds have nuances that may be better described by adjective-noun pairs such as slow car, and verb-noun pairs such as flying insects, which are under explored. Therefore, in this work we investigate the relation between audio content and both adjective-noun pairs and verb-noun pairs. Due to the lack of datasets with these kinds of annotations, we collected and processed the AudioPairBank corpus consisting of a combined total of 1,123 pairs and over 33,000 audio files. One contribution is the previously unavailable documentation of the challenges and implications of collecting audio recordings with these type of labels. A second contribution is to show the degree of correlation between the audio content and the labels through sound recognition experiments, which yielded results of 70% accuracy, hence also providing a performance benchmark. The results and study in this paper encourage further exploration of the nuances in audio and are meant to complement similar research performed on images and text in multimedia analysis.
Tasks
Published	2016-07-13
URL	http://arxiv.org/abs/1607.03766v3
PDF	http://arxiv.org/pdf/1607.03766v3.pdf
PWC	https://paperswithcode.com/paper/audiopairbank-towards-a-large-scale-tag-pair
Repo
Framework

Gradient Descent Learns Linear Dynamical Systems


Title	Gradient Descent Learns Linear Dynamical Systems
Authors	Moritz Hardt, Tengyu Ma, Benjamin Recht
Abstract	We prove that stochastic gradient descent efficiently converges to the global optimizer of the maximum likelihood objective of an unknown linear time-invariant dynamical system from a sequence of noisy observations generated by the system. Even though the objective function is non-convex, we provide polynomial running time and sample complexity bounds under strong but natural assumptions. Linear systems identification has been studied for many decades, yet, to the best of our knowledge, these are the first polynomial guarantees for the problem we consider.
Tasks
Published	2016-09-16
URL	http://arxiv.org/abs/1609.05191v2
PDF	http://arxiv.org/pdf/1609.05191v2.pdf
PWC	https://paperswithcode.com/paper/gradient-descent-learns-linear-dynamical
Repo
Framework

Spatial Decompositions for Large Scale SVMs


Title	Spatial Decompositions for Large Scale SVMs
Authors	Philipp Thomann, Ingrid Blaschzyk, Mona Meister, Ingo Steinwart
Abstract	Although support vector machines (SVMs) are theoretically well understood, their underlying optimization problem becomes very expensive, if, for example, hundreds of thousands of samples and a non-linear kernel are considered. Several approaches have been proposed in the past to address this serious limitation. In this work we investigate a decomposition strategy that learns on small, spatially defined data chunks. Our contributions are two fold: On the theoretical side we establish an oracle inequality for the overall learning method using the hinge loss, and show that the resulting rates match those known for SVMs solving the complete optimization problem with Gaussian kernels. On the practical side we compare our approach to learning SVMs on small, randomly chosen chunks. Here it turns out that for comparable training times our approach is significantly faster during testing and also reduces the test error in most cases significantly. Furthermore, we show that our approach easily scales up to 10 million training samples: including hyper-parameter selection using cross validation, the entire training only takes a few hours on a single machine. Finally, we report an experiment on 32 million training samples. All experiments used liquidSVM (Steinwart and Thomann, 2017).
Tasks
Published	2016-12-01
URL	http://arxiv.org/abs/1612.00374v2
PDF	http://arxiv.org/pdf/1612.00374v2.pdf
PWC	https://paperswithcode.com/paper/spatial-decompositions-for-large-scale-svms
Repo
Framework

Natural Language Comprehension with the EpiReader


Title	Natural Language Comprehension with the EpiReader
Authors	Adam Trischler, Zheng Ye, Xingdi Yuan, Kaheer Suleman
Abstract	We present the EpiReader, a novel model for machine comprehension of text. Machine comprehension of unstructured, real-world text is a major research goal for natural language processing. Current tests of machine comprehension pose questions whose answers can be inferred from some supporting text, and evaluate a model’s response to the questions. The EpiReader is an end-to-end neural model comprising two components: the first component proposes a small set of candidate answers after comparing a question to its supporting text, and the second component formulates hypotheses using the proposed candidates and the question, then reranks the hypotheses based on their estimated concordance with the supporting text. We present experiments demonstrating that the EpiReader sets a new state-of-the-art on the CNN and Children’s Book Test machine comprehension benchmarks, outperforming previous neural models by a significant margin.
Tasks	Question Answering, Reading Comprehension
Published	2016-06-07
URL	http://arxiv.org/abs/1606.02270v2
PDF	http://arxiv.org/pdf/1606.02270v2.pdf
PWC	https://paperswithcode.com/paper/natural-language-comprehension-with-the
Repo
Framework

Quantized neural network design under weight capacity constraint


Title	Quantized neural network design under weight capacity constraint
Authors	Sungho Shin, Kyuyeon Hwang, Wonyong Sung
Abstract	The complexity of deep neural network algorithms for hardware implementation can be lowered either by scaling the number of units or reducing the word-length of weights. Both approaches, however, can accompany the performance degradation although many types of research are conducted to relieve this problem. Thus, it is an important question which one, between the network size scaling and the weight quantization, is more effective for hardware optimization. For this study, the performances of fully-connected deep neural networks (FCDNNs) and convolutional neural networks (CNNs) are evaluated while changing the network complexity and the word-length of weights. Based on these experiments, we present the effective compression ratio (ECR) to guide the trade-off between the network size and the precision of weights when the hardware resource is limited.
Tasks	Quantization
Published	2016-11-19
URL	http://arxiv.org/abs/1611.06342v1
PDF	http://arxiv.org/pdf/1611.06342v1.pdf
PWC	https://paperswithcode.com/paper/quantized-neural-network-design-under-weight
Repo
Framework