July 29, 2019

2704 words 13 mins read

Paper Group ANR 18

PBODL : Parallel Bayesian Online Deep Learning for Click-Through Rate Prediction in Tencent Advertising System. Learning Non-Discriminatory Predictors. Automated Identification of Trampoline Skills Using Computer Vision Extracted Pose Estimation. Measuring the Effect of Discourse Relations on Blog Summarization. A First Derivative Potts Model for S …

PBODL : Parallel Bayesian Online Deep Learning for Click-Through Rate Prediction in Tencent Advertising System


Title	PBODL : Parallel Bayesian Online Deep Learning for Click-Through Rate Prediction in Tencent Advertising System
Authors	Xun Liu, Wei Xue, Lei Xiao, Bo Zhang
Abstract	We describe a parallel bayesian online deep learning framework (PBODL) for click-through rate (CTR) prediction within today’s Tencent advertising system, which provides quick and accurate learning of user preferences. We first explain the framework with a deep probit regression model, which is trained with probabilistic back-propagation in the mode of assumed Gaussian density filtering. Then we extend the model family to a variety of bayesian online models with increasing feature embedding capabilities, such as Sparse-MLP, FM-MLP and FFM-MLP. Finally, we implement a parallel training system based on a stream computing infrastructure and parameter servers. Experiments with public available datasets and Tencent industrial datasets show that models within our framework perform better than several common online models, such as AdPredictor, FTRL-Proximal and MatchBox. Online A/B test within Tencent advertising system further proves that our framework could achieve CTR and CPM lift by learning more quickly and accurately.
Tasks	Click-Through Rate Prediction
Published	2017-07-04
URL	http://arxiv.org/abs/1707.00802v2
PDF	http://arxiv.org/pdf/1707.00802v2.pdf
PWC	https://paperswithcode.com/paper/pbodl-parallel-bayesian-online-deep-learning
Repo
Framework

Learning Non-Discriminatory Predictors


Title	Learning Non-Discriminatory Predictors
Authors	Blake Woodworth, Suriya Gunasekar, Mesrob I. Ohannessian, Nathan Srebro
Abstract	We consider learning a predictor which is non-discriminatory with respect to a “protected attribute” according to the notion of “equalized odds” proposed by Hardt et al. [2016]. We study the problem of learning such a non-discriminatory predictor from a finite training set, both statistically and computationally. We show that a post-hoc correction approach, as suggested by Hardt et al, can be highly suboptimal, present a nearly-optimal statistical procedure, argue that the associated computational problem is intractable, and suggest a second moment relaxation of the non-discrimination definition for which learning is tractable.
Tasks
Published	2017-02-20
URL	http://arxiv.org/abs/1702.06081v3
PDF	http://arxiv.org/pdf/1702.06081v3.pdf
PWC	https://paperswithcode.com/paper/learning-non-discriminatory-predictors
Repo
Framework

Automated Identification of Trampoline Skills Using Computer Vision Extracted Pose Estimation


Title	Automated Identification of Trampoline Skills Using Computer Vision Extracted Pose Estimation
Authors	Paul W. Connolly, Guenole C. Silvestre, Chris J. Bleakley
Abstract	A novel method to identify trampoline skills using a single video camera is proposed herein. Conventional computer vision techniques are used for identification, estimation, and tracking of the gymnast’s body in a video recording of the routine. For each frame, an open source convolutional neural network is used to estimate the pose of the athlete’s body. Body orientation and joint angle estimates are extracted from these pose estimates. The trajectories of these angle estimates over time are compared with those of labelled reference skills. A nearest neighbour classifier utilising a mean squared error distance metric is used to identify the skill performed. A dataset containing 714 skill examples with 20 distinct skills performed by adult male and female gymnasts was recorded and used for evaluation of the system. The system was found to achieve a skill identification accuracy of 80.7% for the dataset.
Tasks	Pose Estimation
Published	2017-09-11
URL	http://arxiv.org/abs/1709.03399v1
PDF	http://arxiv.org/pdf/1709.03399v1.pdf
PWC	https://paperswithcode.com/paper/automated-identification-of-trampoline-skills
Repo
Framework

Measuring the Effect of Discourse Relations on Blog Summarization


Title	Measuring the Effect of Discourse Relations on Blog Summarization
Authors	Shamima Mithun, Leila Kosseim
Abstract	The work presented in this paper attempts to evaluate and quantify the use of discourse relations in the context of blog summarization and compare their use to more traditional and factual texts. Specifically, we measured the usefulness of 6 discourse relations - namely comparison, contingency, illustration, attribution, topic-opinion, and attributive for the task of text summarization from blogs. We have evaluated the effect of each relation using the TAC 2008 opinion summarization dataset and compared them with the results with the DUC 2007 dataset. The results show that in both textual genres, contingency, comparison, and illustration relations provide a significant improvement on summarization content; while attribution, topic-opinion, and attributive relations do not provide a consistent and significant improvement. These results indicate that, at least for summarization, discourse relations are just as useful for informal and affective texts as for more traditional news articles.
Tasks	Text Summarization
Published	2017-08-19
URL	http://arxiv.org/abs/1708.05803v1
PDF	http://arxiv.org/pdf/1708.05803v1.pdf
PWC	https://paperswithcode.com/paper/measuring-the-effect-of-discourse-relations
Repo
Framework

A First Derivative Potts Model for Segmentation and Denoising Using ILP


Title	A First Derivative Potts Model for Segmentation and Denoising Using ILP
Authors	Ruobing Shen, Gerhard Reinelt, Stéphane Canu
Abstract	Unsupervised image segmentation and denoising are two fundamental tasks in image processing. Usually, graph based models such as multicut are used for segmentation and variational models are employed for denoising. Our approach addresses both problems at the same time. We propose a novel ILP formulation of the first derivative Potts model with the $\ell_1$ data term, where binary variables are introduced to deal with the $\ell_0$ norm of the regularization term. The ILP is then solved by a standard off-the-shelf MIP solver. Numerical experiments are compared with the multicut problem.
Tasks	Denoising, Semantic Segmentation
Published	2017-09-21
URL	http://arxiv.org/abs/1709.07212v2
PDF	http://arxiv.org/pdf/1709.07212v2.pdf
PWC	https://paperswithcode.com/paper/a-first-derivative-potts-model-for
Repo
Framework

An overview and comparative analysis of Recurrent Neural Networks for Short Term Load Forecasting


Title	An overview and comparative analysis of Recurrent Neural Networks for Short Term Load Forecasting
Authors	Filippo Maria Bianchi, Enrico Maiorino, Michael C. Kampffmeyer, Antonello Rizzi, Robert Jenssen
Abstract	The key component in forecasting demand and consumption of resources in a supply network is an accurate prediction of real-valued time series. Indeed, both service interruptions and resource waste can be reduced with the implementation of an effective forecasting system. Significant research has thus been devoted to the design and development of methodologies for short term load forecasting over the past decades. A class of mathematical models, called Recurrent Neural Networks, are nowadays gaining renewed interest among researchers and they are replacing many practical implementation of the forecasting systems, previously based on static methods. Despite the undeniable expressive power of these architectures, their recurrent nature complicates their understanding and poses challenges in the training procedures. Recently, new important families of recurrent architectures have emerged and their applicability in the context of load forecasting has not been investigated completely yet. In this paper we perform a comparative study on the problem of Short-Term Load Forecast, by using different classes of state-of-the-art Recurrent Neural Networks. We test the reviewed models first on controlled synthetic tasks and then on different real datasets, covering important practical cases of study. We provide a general overview of the most important architectures and we define guidelines for configuring the recurrent networks to predict real-valued time series.
Tasks	Load Forecasting, Time Series
Published	2017-05-11
URL	http://arxiv.org/abs/1705.04378v2
PDF	http://arxiv.org/pdf/1705.04378v2.pdf
PWC	https://paperswithcode.com/paper/an-overview-and-comparative-analysis-of
Repo
Framework

Identifying Mild Traumatic Brain Injury Patients From MR Images Using Bag of Visual Words


Title	Identifying Mild Traumatic Brain Injury Patients From MR Images Using Bag of Visual Words
Authors	Shervin Minaee, Siyun Wang, Yao Wang, Sohae Chung, Xiuyuan Wang, Els Fieremans, Steven Flanagan, Joseph Rath, Yvonne W. Lui
Abstract	Mild traumatic brain injury (mTBI) is a growing public health problem with an estimated incidence of one million people annually in US. Neurocognitive tests are used to both assess the patient condition and to monitor the patient progress. This work aims to directly use MR images taken shortly after injury to detect whether a patient suffers from mTBI, by incorporating machine learning and computer vision techniques to learn features suitable discriminating between mTBI and normal patients. We focus on 3 regions in brain, and extract multiple patches from them, and use bag-of-visual-word technique to represent each subject as a histogram of representative patterns derived from patches from all training subjects. After extracting the features, we use greedy forward feature selection, to choose a subset of features which achieves highest accuracy. We show through experimental studies that BoW features perform better than the simple mean value features which were used previously.
Tasks	Feature Selection
Published	2017-10-18
URL	http://arxiv.org/abs/1710.06824v3
PDF	http://arxiv.org/pdf/1710.06824v3.pdf
PWC	https://paperswithcode.com/paper/identifying-mild-traumatic-brain-injury
Repo
Framework

Privacy with Estimation Guarantees


Title	Privacy with Estimation Guarantees
Authors	Hao Wang, Lisa Vo, Flavio P. Calmon, Muriel Médard, Ken R. Duffy, Mayank Varia
Abstract	We study the central problem in data privacy: how to share data with an analyst while providing both privacy and utility guarantees to the user that owns the data. In this setting, we present an estimation-theoretic analysis of the privacy-utility trade-off (PUT). Here, an analyst is allowed to reconstruct (in a mean-squared error sense) certain functions of the data (utility), while other private functions should not be reconstructed with distortion below a certain threshold (privacy). We demonstrate how chi-square information captures the fundamental PUT in this case and provide bounds for the best PUT. We propose a convex program to compute privacy-assuring mappings when the functions to be disclosed and hidden are known a priori and the data distribution is known. We derive lower bounds on the minimum mean-squared error of estimating a target function from the disclosed data and evaluate the robustness of our approach when an empirical distribution is used to compute the privacy-assuring mappings instead of the true data distribution. We illustrate the proposed approach through two numerical experiments.
Tasks
Published	2017-10-02
URL	https://arxiv.org/abs/1710.00447v5
PDF	https://arxiv.org/pdf/1710.00447v5.pdf
PWC	https://paperswithcode.com/paper/privacy-with-estimation-guarantees
Repo
Framework

Harvesting Creative Templates for Generating Stylistically Varied Restaurant Reviews


Title	Harvesting Creative Templates for Generating Stylistically Varied Restaurant Reviews
Authors	Shereen Oraby, Sheideh Homayon, Marilyn Walker
Abstract	Many of the creative and figurative elements that make language exciting are lost in translation in current natural language generation engines. In this paper, we explore a method to harvest templates from positive and negative reviews in the restaurant domain, with the goal of vastly expanding the types of stylistic variation available to the natural language generator. We learn hyperbolic adjective patterns that are representative of the strongly-valenced expressive language commonly used in either positive or negative reviews. We then identify and delexicalize entities, and use heuristics to extract generation templates from review sentences. We evaluate the learned templates against more traditional review templates, using subjective measures of “convincingness”, “interestingness”, and “naturalness”. Our results show that the learned templates score highly on these measures. Finally, we analyze the linguistic categories that characterize the learned positive and negative templates. We plan to use the learned templates to improve the conversational style of dialogue systems in the restaurant domain.
Tasks	Text Generation
Published	2017-09-15
URL	http://arxiv.org/abs/1709.05308v1
PDF	http://arxiv.org/pdf/1709.05308v1.pdf
PWC	https://paperswithcode.com/paper/harvesting-creative-templates-for-generating
Repo
Framework

Who’s to say what’s funny? A computer using Language Models and Deep Learning, That’s Who!


Title	Who’s to say what’s funny? A computer using Language Models and Deep Learning, That’s Who!
Authors	Xinru Yan, Ted Pedersen
Abstract	Humor is a defining characteristic of human beings. Our goal is to develop methods that automatically detect humorous statements and rank them on a continuous scale. In this paper we report on results using a Language Model approach, and outline our plans for using methods from Deep Learning.
Tasks	Language Modelling
Published	2017-05-29
URL	http://arxiv.org/abs/1705.10272v1
PDF	http://arxiv.org/pdf/1705.10272v1.pdf
PWC	https://paperswithcode.com/paper/whos-to-say-whats-funny-a-computer-using
Repo
Framework

Feature Enhancement Network: A Refined Scene Text Detector


Title	Feature Enhancement Network: A Refined Scene Text Detector
Authors	Sheng Zhang, Yuliang Liu, Lianwen Jin, Canjie Luo
Abstract	In this paper, we propose a refined scene text detector with a \textit{novel} Feature Enhancement Network (FEN) for Region Proposal and Text Detection Refinement. Retrospectively, both region proposal with \textit{only} $3\times 3$ sliding-window feature and text detection refinement with \textit{single scale} high level feature are insufficient, especially for smaller scene text. Therefore, we design a new FEN network with \textit{task-specific}, \textit{low} and \textit{high} level semantic features fusion to improve the performance of text detection. Besides, since \textit{unitary} position-sensitive RoI pooling in general object detection is unreasonable for variable text regions, an \textit{adaptively weighted} position-sensitive RoI pooling layer is devised for further enhancing the detecting accuracy. To tackle the \textit{sample-imbalance} problem during the refinement stage, we also propose an effective \textit{positives mining} strategy for efficiently training our network. Experiments on ICDAR 2011 and 2013 robust text detection benchmarks demonstrate that our method can achieve state-of-the-art results, outperforming all reported methods in terms of F-measure.
Tasks	Object Detection
Published	2017-11-12
URL	http://arxiv.org/abs/1711.04249v1
PDF	http://arxiv.org/pdf/1711.04249v1.pdf
PWC	https://paperswithcode.com/paper/feature-enhancement-network-a-refined-scene
Repo
Framework

DNN-Buddies: A Deep Neural Network-Based Estimation Metric for the Jigsaw Puzzle Problem


Title	DNN-Buddies: A Deep Neural Network-Based Estimation Metric for the Jigsaw Puzzle Problem
Authors	Dror Sholomon, Eli David, Nathan S. Netanyahu
Abstract	This paper introduces the first deep neural network-based estimation metric for the jigsaw puzzle problem. Given two puzzle piece edges, the neural network predicts whether or not they should be adjacent in the correct assembly of the puzzle, using nothing but the pixels of each piece. The proposed metric exhibits an extremely high precision even though no manual feature extraction is performed. When incorporated into an existing puzzle solver, the solution’s accuracy increases significantly, achieving thereby a new state-of-the-art standard.
Tasks
Published	2017-11-23
URL	http://arxiv.org/abs/1711.08762v1
PDF	http://arxiv.org/pdf/1711.08762v1.pdf
PWC	https://paperswithcode.com/paper/dnn-buddies-a-deep-neural-network-based
Repo
Framework

Opinion Recommendation using Neural Memory Model


Title	Opinion Recommendation using Neural Memory Model
Authors	Zhongqing Wang, Yue Zhang
Abstract	We present opinion recommendation, a novel task of jointly predicting a custom review with a rating score that a certain user would give to a certain product or service, given existing reviews and rating scores to the product or service by other users, and the reviews that the user has given to other products and services. A characteristic of opinion recommendation is the reliance of multiple data sources for multi-task joint learning, which is the strength of neural models. We use a single neural network to model users and products, capturing their correlation and generating customised product representations using a deep memory network, from which customised ratings and reviews are constructed jointly. Results show that our opinion recommendation system gives ratings that are closer to real user ratings on Yelp.com data compared with Yelp’s own ratings, and our methods give better results compared to several pipelines baselines using state-of-the-art sentiment rating and summarization systems.
Tasks
Published	2017-02-06
URL	http://arxiv.org/abs/1702.01517v1
PDF	http://arxiv.org/pdf/1702.01517v1.pdf
PWC	https://paperswithcode.com/paper/opinion-recommendation-using-neural-memory
Repo
Framework

Denoising Linear Models with Permuted Data


Title	Denoising Linear Models with Permuted Data
Authors	Ashwin Pananjady, Martin J. Wainwright, Thomas A. Courtade
Abstract	The multivariate linear regression model with shuffled data and additive Gaussian noise arises in various correspondence estimation and matching problems. Focusing on the denoising aspect of this problem, we provide a characterization the minimax error rate that is sharp up to logarithmic factors. We also analyze the performance of two versions of a computationally efficient estimator, and establish their consistency for a large range of input parameters. Finally, we provide an exact algorithm for the noiseless problem and demonstrate its performance on an image point-cloud matching task. Our analysis also extends to datasets with outliers.
Tasks	Denoising
Published	2017-04-24
URL	http://arxiv.org/abs/1704.07461v1
PDF	http://arxiv.org/pdf/1704.07461v1.pdf
PWC	https://paperswithcode.com/paper/denoising-linear-models-with-permuted-data
Repo
Framework

Robust Stackelberg Equilibria in Extensive-Form Games and Extension to Limited Lookahead


Title	Robust Stackelberg Equilibria in Extensive-Form Games and Extension to Limited Lookahead
Authors	Christian Kroer, Gabriele Farina, Tuomas Sandholm
Abstract	Stackelberg equilibria have become increasingly important as a solution concept in computational game theory, largely inspired by practical problems such as security settings. In practice, however, there is typically uncertainty regarding the model about the opponent. This paper is, to our knowledge, the first to investigate Stackelberg equilibria under uncertainty in extensive-form games, one of the broadest classes of game. We introduce robust Stackelberg equilibria, where the uncertainty is about the opponent’s payoffs, as well as ones where the opponent has limited lookahead and the uncertainty is about the opponent’s node evaluation function. We develop a new mixed-integer program for the deterministic limited-lookahead setting. We then extend the program to the robust setting for Stackelberg equilibrium under unlimited and under limited lookahead by the opponent. We show that for the specific case of interval uncertainty about the opponent’s payoffs (or about the opponent’s node evaluations in the case of limited lookahead), robust Stackelberg equilibria can be computed with a mixed-integer program that is of the same asymptotic size as that for the deterministic setting.
Tasks
Published	2017-11-21
URL	http://arxiv.org/abs/1711.08080v1
PDF	http://arxiv.org/pdf/1711.08080v1.pdf
PWC	https://paperswithcode.com/paper/robust-stackelberg-equilibria-in-extensive
Repo
Framework