October 16, 2019

2698 words 13 mins read

Paper Group NANR 7

Paper Group NANR 7

Cross-language Article Linking Using Cross-Encyclopedia Entity Embedding. Topkapi: Parallel and Fast Sketches for Finding Top-K Frequent Elements. Estimation of cross-lingual news similarities using text-mining methods. Which Scores to Predict in Sentence Regression for Text Summarization?. Dividing and Aggregating Network for Multi-view Action Rec …

Cross-language Article Linking Using Cross-Encyclopedia Entity Embedding

Title Cross-language Article Linking Using Cross-Encyclopedia Entity Embedding
Authors Chun-Kai Wu, Richard Tzong-Han Tsai
Abstract Cross-language article linking (CLAL) is the task of finding corresponding article pairs of different languages across encyclopedias. This task is a difficult disambiguation problem in which one article must be selected among several candidate articles with similar titles and contents. Existing works focus on engineering text-based or link-based features for this task, which is a time-consuming job, and some of these features are only applicable within the same encyclopedia. In this paper, we address these problems by proposing cross-encyclopedia entity embedding. Unlike other works, our proposed method does not rely on known cross-language pairs. We apply our method to CLAL between English Wikipedia and Chinese Baidu Baike. Our features improve performance relative to the baseline by 29.62{%}. Tested 30 times, our system achieved an average improvement of 2.76{%} over the current best system (26.86{%} over baseline), a statistically significant result.
Tasks Information Retrieval
Published 2018-06-01
URL https://www.aclweb.org/anthology/N18-2054/
PDF https://www.aclweb.org/anthology/N18-2054
PWC https://paperswithcode.com/paper/cross-language-article-linking-using-cross
Repo
Framework

Topkapi: Parallel and Fast Sketches for Finding Top-K Frequent Elements

Title Topkapi: Parallel and Fast Sketches for Finding Top-K Frequent Elements
Authors Ankush Mandal, He Jiang, Anshumali Shrivastava, Vivek Sarkar
Abstract Identifying the top-K frequent items is one of the most common and important operations in large data processing systems. As a result, several solutions have been proposed to solve this problem approximately. In this paper, we identify that in modern distributed settings with both multi-node as well as multi-core parallelism, existing algorithms, although theoretically sound, are suboptimal from the performance perspective. In particular, for identifying top-K frequent items, Count-Min Sketch (CMS) has fantastic update time but lack the important property of reducibility which is needed for exploiting available massive data parallelism. On the other end, popular Frequent algorithm (FA) leads to reducible summaries but the update costs are significant. In this paper, we present Topkapi, a fast and parallel algorithm for finding top-K frequent items, which gives the best of both worlds, i.e., it is reducible as well as efficient update time similar to CMS. Topkapi possesses strong theoretical guarantees and leads to significant performance gains due to increased parallelism, relative to past work.
Tasks
Published 2018-12-01
URL http://papers.nips.cc/paper/8287-topkapi-parallel-and-fast-sketches-for-finding-top-k-frequent-elements
PDF http://papers.nips.cc/paper/8287-topkapi-parallel-and-fast-sketches-for-finding-top-k-frequent-elements.pdf
PWC https://paperswithcode.com/paper/topkapi-parallel-and-fast-sketches-for
Repo
Framework

Estimation of cross-lingual news similarities using text-mining methods

Title Estimation of cross-lingual news similarities using text-mining methods
Authors Zhouhao Wang, Enda Liu, Hiroki Sakaji, Tomoki Ito, Kiyoshi Izumi, Kota Tsubouchi, Tatsuo Yamashita
Abstract Every second, innumerable text data, including all kinds news, reports, messages, reviews, comments, and twits have been generated on the Internet, which is written not only in English but also in other languages such as Chinese, Japanese, French and so on. Not only SNS sites but also worldwide news agency such as Thomson Reuters News provide news reported in more than 20 languages, reflecting the significance of the multilingual information. In this research, by taking advantage of multi-lingual text resources provided by the Thomson Reuters News, we developed a bidirectional LSTM based method to calculate cross-lingual semantic text similarity for long text and short text respectively. Thus, users could understand the situation comprehensively, by investigating similar and related cross-lingual articles, when there an important news comes in.
Tasks
Published 2018-01-01
URL https://openreview.net/forum?id=r1HNP0eCW
PDF https://openreview.net/pdf?id=r1HNP0eCW
PWC https://paperswithcode.com/paper/estimation-of-cross-lingual-news-similarities
Repo
Framework

Which Scores to Predict in Sentence Regression for Text Summarization?

Title Which Scores to Predict in Sentence Regression for Text Summarization?
Authors Markus Zopf, Eneldo Loza Menc{'\i}a, Johannes F{"u}rnkranz
Abstract The task of automatic text summarization is to generate a short text that summarizes the most important information in a given set of documents. Sentence regression is an emerging branch in automatic text summarizations. Its key idea is to estimate the importance of information via learned utility scores for individual sentences. These scores are then used for selecting sentences from the source documents, typically according to a greedy selection strategy. Recently proposed state-of-the-art models learn to predict ROUGE recall scores of individual sentences, which seems reasonable since the final summaries are evaluated according to ROUGE recall. In this paper, we show in extensive experiments that following this intuition leads to suboptimal results and that learning to predict ROUGE precision scores leads to better results. The crucial difference is to aim not at covering as much information as possible but at wasting as little space as possible in every greedy step.
Tasks Text Summarization
Published 2018-06-01
URL https://www.aclweb.org/anthology/N18-1161/
PDF https://www.aclweb.org/anthology/N18-1161
PWC https://paperswithcode.com/paper/which-scores-to-predict-in-sentence
Repo
Framework

Dividing and Aggregating Network for Multi-view Action Recognition

Title Dividing and Aggregating Network for Multi-view Action Recognition
Authors Dongang Wang, Wanli Ouyang, Wen Li, Dong Xu
Abstract In this paper, we propose a new Dividing and Aggregating Network (DA-Net) for multi-view action recognition. In our DA-Net, we learn view-independent representations shared by all views at lower layers, while we learn one view-specific representation for each view at higher layers. We then train view-specific action classifiers based on the view-specific representation for each view and a view classifier based on the shared representation at lower layers. The view classifier is used to predict how likely each video belongs to each view. Finally, the predicted view probabilities from multiple views are used as the weights when fusing the prediction scores of view-specific action classifiers. We also propose a new approach based on the conditional random field (CRF) formulation to pass message among view-specific representations from different branches to help each other. Comprehensive experiments on two benchmark datasets clearly demonstrate the effectiveness of our proposed DA-Net for multi-view action recognition.
Tasks Temporal Action Localization
Published 2018-09-01
URL http://openaccess.thecvf.com/content_ECCV_2018/html/Dongang_Wang_Dividing_and_Aggregating_ECCV_2018_paper.html
PDF http://openaccess.thecvf.com/content_ECCV_2018/papers/Dongang_Wang_Dividing_and_Aggregating_ECCV_2018_paper.pdf
PWC https://paperswithcode.com/paper/dividing-and-aggregating-network-for-multi
Repo
Framework

The Cluster Description Problem - Complexity Results, Formulations and Approximations

Title The Cluster Description Problem - Complexity Results, Formulations and Approximations
Authors Ian Davidson, Antoine Gourru, S Ravi
Abstract Consider the situation where you are given an existing $k$-way clustering $\pi$. A challenge for explainable AI is to find a compact and distinct explanations of each cluster which in this paper is using instance-level descriptors/tags from a common dictionary. Since the descriptors/tags were not given to the clustering method, this is not a semi-supervised learning situation. We show that the \emph{feasibility} problem of just testing whether any distinct description (not the most compact) exists is generally intractable for just two clusters. This means that unless \textbf{P} = \cnp, there cannot exist an efficient algorithm for the cluster description problem. Hence, we explore ILP formulations for smaller problems and a relaxed but restricted setting that leads to a polynomial time algorithm for larger problems. We explore several extension to the basic setting such as the ability to ignore some instances and composition constraints on the descriptions of the clusters. We show our formulation’s usefulness on Twitter data where the communities were found using social connectivity (i.e. \texttt{follower} relation) but the explanation of the communities is based on behavioral properties of the nodes (i.e. hashtag usage) not available to the clustering method.
Tasks
Published 2018-12-01
URL http://papers.nips.cc/paper/7857-the-cluster-description-problem-complexity-results-formulations-and-approximations
PDF http://papers.nips.cc/paper/7857-the-cluster-description-problem-complexity-results-formulations-and-approximations.pdf
PWC https://paperswithcode.com/paper/the-cluster-description-problem-complexity
Repo
Framework

Multi-Class Learning: From Theory to Algorithm

Title Multi-Class Learning: From Theory to Algorithm
Authors Jian Li, Yong Liu, Rong Yin, Hua Zhang, Lizhong Ding, Weiping Wang
Abstract In this paper, we study the generalization performance of multi-class classification and obtain a shaper data-dependent generalization error bound with fast convergence rate, substantially improving the state-of-art bounds in the existing data-dependent generalization analysis. The theoretical analysis motivates us to devise two effective multi-class kernel learning algorithms with statistical guarantees. Experimental results show that our proposed methods can significantly outperform the existing multi-class classification methods.
Tasks
Published 2018-12-01
URL http://papers.nips.cc/paper/7431-multi-class-learning-from-theory-to-algorithm
PDF http://papers.nips.cc/paper/7431-multi-class-learning-from-theory-to-algorithm.pdf
PWC https://paperswithcode.com/paper/multi-class-learning-from-theory-to-algorithm
Repo
Framework

Personalized Text Retrieval for Learners of Chinese as a Foreign Language

Title Personalized Text Retrieval for Learners of Chinese as a Foreign Language
Authors Chak Yan Yeung, John Lee
Abstract This paper describes a personalized text retrieval algorithm that helps language learners select the most suitable reading material in terms of vocabulary complexity. The user first rates their knowledge of a small set of words, chosen by a graph-based active learning model. The system trains a complex word identification model on this set, and then applies the model to find texts that contain the desired proportion of new, challenging, and familiar vocabulary. In an evaluation on learners of Chinese as a foreign language, we show that this algorithm is effective in identifying simpler texts for low-proficiency learners, and more challenging ones for high-proficiency learners.
Tasks Active Learning, Complex Word Identification
Published 2018-08-01
URL https://www.aclweb.org/anthology/C18-1292/
PDF https://www.aclweb.org/anthology/C18-1292
PWC https://paperswithcode.com/paper/personalized-text-retrieval-for-learners-of
Repo
Framework

Learning Versatile Filters for Efficient Convolutional Neural Networks

Title Learning Versatile Filters for Efficient Convolutional Neural Networks
Authors Yunhe Wang, Chang Xu, Chunjing Xu, Chao Xu, Dacheng Tao
Abstract This paper introduces versatile filters to construct efficient convolutional neural network. Considering the demands of efficient deep learning techniques running on cost-effective hardware, a number of methods have been developed to learn compact neural networks. Most of these works aim to slim down filters in different ways, e.g., investigating small, sparse or binarized filters. In contrast, we treat filters from an additive perspective. A series of secondary filters can be derived from a primary filter. These secondary filters all inherit in the primary filter without occupying more storage, but once been unfolded in computation they could significantly enhance the capability of the filter by integrating information extracted from different receptive fields. Besides spatial versatile filters, we additionally investigate versatile filters from the channel perspective. The new techniques are general to upgrade filters in existing CNNs. Experimental results on benchmark datasets and neural networks demonstrate that CNNs constructed with our versatile filters are able to achieve comparable accuracy as that of original filters, but require less memory and FLOPs.
Tasks
Published 2018-12-01
URL http://papers.nips.cc/paper/7433-learning-versatile-filters-for-efficient-convolutional-neural-networks
PDF http://papers.nips.cc/paper/7433-learning-versatile-filters-for-efficient-convolutional-neural-networks.pdf
PWC https://paperswithcode.com/paper/learning-versatile-filters-for-efficient
Repo
Framework

Accelerated Stochastic Matrix Inversion: General Theory and Speeding up BFGS Rules for Faster Second-Order Optimization

Title Accelerated Stochastic Matrix Inversion: General Theory and Speeding up BFGS Rules for Faster Second-Order Optimization
Authors Robert Gower, Filip Hanzely, Peter Richtarik, Sebastian U. Stich
Abstract We present the first accelerated randomized algorithm for solving linear systems in Euclidean spaces. One essential problem of this type is the matrix inversion problem. In particular, our algorithm can be specialized to invert positive definite matrices in such a way that all iterates (approximate solutions) generated by the algorithm are positive definite matrices themselves. This opens the way for many applications in the field of optimization and machine learning. As an application of our general theory, we develop the first accelerated (deterministic and stochastic) quasi-Newton updates. Our updates lead to provably more aggressive approximations of the inverse Hessian, and lead to speed-ups over classical non-accelerated rules in numerical experiments. Experiments with empirical risk minimization show that our rules can accelerate training of machine learning models.
Tasks
Published 2018-12-01
URL http://papers.nips.cc/paper/7434-accelerated-stochastic-matrix-inversion-general-theory-and-speeding-up-bfgs-rules-for-faster-second-order-optimization
PDF http://papers.nips.cc/paper/7434-accelerated-stochastic-matrix-inversion-general-theory-and-speeding-up-bfgs-rules-for-faster-second-order-optimization.pdf
PWC https://paperswithcode.com/paper/accelerated-stochastic-matrix-inversion
Repo
Framework

DeepPavlov: Open-Source Library for Dialogue Systems

Title DeepPavlov: Open-Source Library for Dialogue Systems
Authors Mikhail Burtsev, Alex Seliverstov, er, Rafael Airapetyan, Mikhail Arkhipov, Dilyara Baymurzina, Nickolay Bushkov, Olga Gureenkova, Taras Khakhulin, Yuri Kuratov, Denis Kuznetsov, Alexey Litinsky, Varvara Logacheva, Alexey Lymar, Valentin Malykh, Maxim Petrov, Vadim Polulyakh, Leonid Pugachev, Alexey Sorokin, Maria Vikhreva, Marat Zaynutdinov
Abstract Adoption of messaging communication and voice assistants has grown rapidly in the last years. This creates a demand for tools that speed up prototyping of feature-rich dialogue systems. An open-source library DeepPavlov is tailored for development of conversational agents. The library prioritises efficiency, modularity, and extensibility with the goal to make it easier to develop dialogue systems from scratch and with limited data available. It supports modular as well as end-to-end approaches to implementation of conversational agents. Conversational agent consists of skills and every skill can be decomposed into components. Components are usually models which solve typical NLP tasks such as intent classification, named entity recognition or pre-trained word vectors. Sequence-to-sequence chit-chat skill, question answering skill or task-oriented skill can be assembled from components provided in the library.
Tasks Intent Classification, Named Entity Recognition, Question Answering
Published 2018-07-01
URL https://www.aclweb.org/anthology/P18-4021/
PDF https://www.aclweb.org/anthology/P18-4021
PWC https://paperswithcode.com/paper/deeppavlov-open-source-library-for-dialogue
Repo
Framework

Understanding and Exploiting the Low-Rank Structure of Deep Networks

Title Understanding and Exploiting the Low-Rank Structure of Deep Networks
Authors Craig Bakker, Michael J. Henry, Nathan O. Hodas
Abstract Training methods for deep networks are primarily variants on stochastic gradient descent. Techniques that use (approximate) second-order information are rarely used because of the computational cost and noise associated with those approaches in deep learning contexts. However, in this paper, we show how feedforward deep networks exhibit a low-rank derivative structure. This low-rank structure makes it possible to use second-order information without needing approximations and without incurring a significantly greater computational cost than gradient descent. To demonstrate this capability, we implement Cubic Regularization (CR) on a feedforward deep network with stochastic gradient descent and two of its variants. There, we use CR to calculate learning rates on a per-iteration basis while training on the MNIST and CIFAR-10 datasets. CR proved particularly successful in escaping plateau regions of the objective function. We also found that this approach requires less problem-specific information (e.g. an optimal initial learning rate) than other first-order methods in order to perform well.
Tasks
Published 2018-01-01
URL https://openreview.net/forum?id=ByJ7obb0b
PDF https://openreview.net/pdf?id=ByJ7obb0b
PWC https://paperswithcode.com/paper/understanding-and-exploiting-the-low-rank
Repo
Framework

Automatically Selecting the Best Dependency Annotation Design with Dynamic Oracles

Title Automatically Selecting the Best Dependency Annotation Design with Dynamic Oracles
Authors Guillaume Wisniewski, Oph{'e}lie Lacroix, Fran{\c{c}}ois Yvon
Abstract This work introduces a new strategy to compare the numerous conventions that have been proposed over the years for expressing dependency structures and discover the one for which a parser will achieve the highest parsing performance. Instead of associating each sentence in the training set with a single gold reference we propose to consider a set of references encoding alternative syntactic representations. Training a parser with a dynamic oracle will then automatically select among all alternatives the reference that will be predicted with the highest accuracy. Experiments on the UD corpora show the validity of this approach.
Tasks
Published 2018-06-01
URL https://www.aclweb.org/anthology/N18-2064/
PDF https://www.aclweb.org/anthology/N18-2064
PWC https://paperswithcode.com/paper/automatically-selecting-the-best-dependency
Repo
Framework

KIPPI: KInetic Polygonal Partitioning of Images

Title KIPPI: KInetic Polygonal Partitioning of Images
Authors Jean-Philippe Bauchet, Florent Lafarge
Abstract Recent works showed that floating polygons can be an interesting alternative to traditional superpixels, especially for analyzing scenes with strong geometric signatures, as man-made environments. Existing algorithms produce homogeneously-sized polygons that fail to capture thin geometric structures and over-partition large uniform areas. We propose a kinetic approach that brings more flexibility on polygon shape and size. The key idea consists in progressively extending pre-detected line-segments until they meet each other. Our experiments demonstrate that output partitions both contain less polygons and better capture geometric structures than those delivered by existing methods. We also show the applicative potential of the method when used as preprocessing in object contouring.
Tasks
Published 2018-06-01
URL http://openaccess.thecvf.com/content_cvpr_2018/html/Bauchet_KIPPI_KInetic_Polygonal_CVPR_2018_paper.html
PDF http://openaccess.thecvf.com/content_cvpr_2018/papers/Bauchet_KIPPI_KInetic_Polygonal_CVPR_2018_paper.pdf
PWC https://paperswithcode.com/paper/kippi-kinetic-polygonal-partitioning-of
Repo
Framework

Video Object Segmentation by Learning Location-Sensitive Embeddings

Title Video Object Segmentation by Learning Location-Sensitive Embeddings
Authors Hai Ci, Chunyu Wang, Yizhou Wang
Abstract We address the problem of video object segmentation which outputs the masks of a target object throughout a video given only a bounding box in the first frame. There are two main challenges for this task. First, the background may contain similar objects as the target. Second, the appearance of the target object may change drastically over time. To tackle these challenges, we propose an end-to-end training network which accomplishes foreground predictions by leveraging the location-sensitive embeddings which are capable to distinguish the pixels of similar objects. To deal with appearance changes, for a test video, we propose a robust model adaptation method which pre-scans the whole video, generates pseudo foreground/background labels and retrains the model based on the labels. Our method outperforms the state-of-the-art methods on the DAVIS and the SegTrack V2 datasets.
Tasks Semantic Segmentation, Video Object Segmentation, Video Semantic Segmentation
Published 2018-09-01
URL http://openaccess.thecvf.com/content_ECCV_2018/html/Hai_Ci_Video_Object_Segmentation_ECCV_2018_paper.html
PDF http://openaccess.thecvf.com/content_ECCV_2018/papers/Hai_Ci_Video_Object_Segmentation_ECCV_2018_paper.pdf
PWC https://paperswithcode.com/paper/video-object-segmentation-by-learning
Repo
Framework
comments powered by Disqus