October 15, 2019

1747 words 9 mins read

Paper Group NANR 208

Paper Group NANR 208

Improved nearest neighbor search using auxiliary information and priority functions. The UIR Uncertainty Corpus for Chinese: Annotating Chinese Microblog Corpus for Uncertainty Identification from Social Media. Using Crowd Agreement for Wordnet Localization. Knowing the Author by the Company His Words Keep. Training with Growing Sets: A Simple Alte …

Improved nearest neighbor search using auxiliary information and priority functions

Title Improved nearest neighbor search using auxiliary information and priority functions
Authors Omid Keivani, Kaushik Sinha
Abstract Nearest neighbor search using random projection trees has recently been shown to achieve superior performance, in terms of better accuracy while retrieving less number of data points, compared to locality sensitive hashing based methods. However, to achieve acceptable nearest neighbor search accuracy for large scale applications, where number of data points and/or number of features can be very large, it requires users to maintain, store and search through large number of such independent random projection trees, which may be undesirable for many practical applications. To address this issue, in this paper we present different search strategies to improve nearest neighbor search performance of a single random projection tree. Our approach exploits properties of single and multiple random projections, which allows us to store meaningful auxiliary information at internal nodes of a random projection tree as well as to design priority functions to guide the search process that results in improved nearest neighbor search performance. Empirical results on multiple real world datasets show that our proposed method improves the search accuracy of a single tree compared to baseline methods.
Tasks
Published 2018-07-01
URL https://icml.cc/Conferences/2018/Schedule?showEvent=2041
PDF http://proceedings.mlr.press/v80/keivani18a/keivani18a.pdf
PWC https://paperswithcode.com/paper/improved-nearest-neighbor-search-using
Repo
Framework

The UIR Uncertainty Corpus for Chinese: Annotating Chinese Microblog Corpus for Uncertainty Identification from Social Media

Title The UIR Uncertainty Corpus for Chinese: Annotating Chinese Microblog Corpus for Uncertainty Identification from Social Media
Authors Binyang Li, Jun Xiang, Le Chen, Xu Han, Xiaoyan Yu, Ruifeng Xu, Tengjiao Wang, Kam-fai Wong
Abstract
Tasks Question Answering
Published 2018-05-01
URL https://www.aclweb.org/anthology/L18-1078/
PDF https://www.aclweb.org/anthology/L18-1078
PWC https://paperswithcode.com/paper/the-uir-uncertainty-corpus-for-chinese
Repo
Framework

Using Crowd Agreement for Wordnet Localization

Title Using Crowd Agreement for Wordnet Localization
Authors Amarsanaa Ganbold, Altangerel Chagnaa, G{'a}bor Bella
Abstract
Tasks Machine Translation
Published 2018-05-01
URL https://www.aclweb.org/anthology/L18-1074/
PDF https://www.aclweb.org/anthology/L18-1074
PWC https://paperswithcode.com/paper/using-crowd-agreement-for-wordnet
Repo
Framework

Knowing the Author by the Company His Words Keep

Title Knowing the Author by the Company His Words Keep
Authors Armin Hoenen, Niko Schenk
Abstract
Tasks Semantic Textual Similarity, Word Embeddings
Published 2018-05-01
URL https://www.aclweb.org/anthology/L18-1083/
PDF https://www.aclweb.org/anthology/L18-1083
PWC https://paperswithcode.com/paper/knowing-the-author-by-the-company-his-words
Repo
Framework

Training with Growing Sets: A Simple Alternative to Curriculum Learning and Self Paced Learning

Title Training with Growing Sets: A Simple Alternative to Curriculum Learning and Self Paced Learning
Authors Melike Nur Mermer, Mehmet Fatih Amasyali
Abstract Curriculum learning and Self paced learning are popular topics in the machine learning that suggest to put the training samples in order by considering their difficulty levels. Studies in these topics show that starting with a small training set and adding new samples according to difficulty levels improves the learning performance. In this paper we experimented that we can also obtain good results by adding the samples randomly without a meaningful order. We compared our method with classical training, Curriculum learning, Self paced learning and their reverse ordered versions. Results of the statistical tests show that the proposed method is better than classical method and similar with the others. These results point a new training regime that removes the process of difficulty level determination in Curriculum and Self paced learning and as successful as these methods.
Tasks
Published 2018-01-01
URL https://openreview.net/forum?id=SJ1fQYlCZ
PDF https://openreview.net/pdf?id=SJ1fQYlCZ
PWC https://paperswithcode.com/paper/training-with-growing-sets-a-simple
Repo
Framework

Network Iterative Learning for Dynamic Deep Neural Networks via Morphism

Title Network Iterative Learning for Dynamic Deep Neural Networks via Morphism
Authors Tao Wei, Changhu Wang, Chang Wen Chen
Abstract In this research, we present a novel learning scheme called network iterative learning for deep neural networks. Different from traditional optimization algorithms that usually optimize directly on a static objective function, we propose in this work to optimize a dynamic objective function in an iterative fashion capable of adapting its function form when being optimized. The optimization is implemented as a series of intermediate neural net functions that is able to dynamically grow into the targeted neural net objective function. This is done via network morphism so that the network knowledge is fully preserved with each network growth. Experimental results demonstrate that the proposed network iterative learning scheme is able to significantly alleviate the degradation problem. Its effectiveness is verified on diverse benchmark datasets.
Tasks
Published 2018-01-01
URL https://openreview.net/forum?id=SkFvV0yC-
PDF https://openreview.net/pdf?id=SkFvV0yC-
PWC https://paperswithcode.com/paper/network-iterative-learning-for-dynamic-deep
Repo
Framework

One event, many representations. Mapping action concepts through visual features.

Title One event, many representations. Mapping action concepts through visual features.
Authors Aless Panunzi, ro, Lorenzo Gregori, Andrea Amelio Ravelli
Abstract
Tasks
Published 2018-05-01
URL https://www.aclweb.org/anthology/L18-1095/
PDF https://www.aclweb.org/anthology/L18-1095
PWC https://paperswithcode.com/paper/one-event-many-representations-mapping-action
Repo
Framework

Contextualized Usage-Based Material Selection

Title Contextualized Usage-Based Material Selection
Authors Dirk De Hertog, Piet Desmet
Abstract
Tasks Language Acquisition, Semantic Textual Similarity
Published 2018-05-01
URL https://www.aclweb.org/anthology/L18-1053/
PDF https://www.aclweb.org/anthology/L18-1053
PWC https://paperswithcode.com/paper/contextualized-usage-based-material-selection
Repo
Framework

Learning Patch Reconstructability for Accelerating Multi-View Stereo

Title Learning Patch Reconstructability for Accelerating Multi-View Stereo
Authors Alex Poms, Chenglei Wu, Shoou-I Yu, Yaser Sheikh
Abstract We present an approach to accelerate multi-view stereo (MVS) by prioritizing computation on image patches that are likely to produce accurate 3D surface reconstructions. Our key insight is that the accuracy of the surface reconstruction from a given image patch can be predicted significantly faster than performing the actual stereo matching. The intuition is that non-specular, fronto-parallel, in-focus patches are more likely to produce accurate surface reconstructions than highly specular, slanted, blurry patches — and that these properties can be reliably predicted from the image itself. By prioritizing stereo matching on a subset of patches that are highly reconstructable and also cover the 3D surface, we are able to accelerate MVS with minimal reduction in accuracy and completeness. To predict the reconstructability score of an image patch from a single view, we train an image-to-reconstructability neural network: the I2RNet. This reconstructability score enables us to efficiently identify image patches that are likely to provide the most accurate surface estimates before performing stereo matching. We demonstrate that the I2RNet, when trained on the ScanNet dataset, generalizes to the DTU and Tanks and Temples MVS datasets. By using our I2RNet with an existing MVS implementation, we show that our method can achieve more than a 30x speed-up over the baseline with only an minimal loss in completeness.
Tasks Stereo Matching, Stereo Matching Hand
Published 2018-06-01
URL http://openaccess.thecvf.com/content_cvpr_2018/html/Poms_Learning_Patch_Reconstructability_CVPR_2018_paper.html
PDF http://openaccess.thecvf.com/content_cvpr_2018/papers/Poms_Learning_Patch_Reconstructability_CVPR_2018_paper.pdf
PWC https://paperswithcode.com/paper/learning-patch-reconstructability-for
Repo
Framework

SIC-GAN: A Self-Improving Collaborative GAN for Decoding Sketch RNNs

Title SIC-GAN: A Self-Improving Collaborative GAN for Decoding Sketch RNNs
Authors Chi-Chun Chuang, Zheng-Xin Weng, Shan-Hung Wu
Abstract Variational RNNs are proposed to output “creative” sequences. Ideally, a collection of sequences produced by a variational RNN should be of both high quality and high variety. However, existing decoders for variational RNNs suffer from a trade-off between quality and variety. In this paper, we seek to learn a variational RNN that decodes high-quality and high-variety sequences. We propose the Self-Improving Collaborative GAN (SIC-GAN), where there are two generators (variational RNNs) collaborating with each other to output a sequence and aiming to trick the discriminator into believing the sequence is of good quality. By deliberately weakening one generator, we can make another stronger in balancing quality and variety. We conduct experiments using the QuickDraw dataset and the results demonstrate the effectiveness of SIC-GAN empirically.
Tasks
Published 2018-01-01
URL https://openreview.net/forum?id=r1nmx5l0W
PDF https://openreview.net/pdf?id=r1nmx5l0W
PWC https://paperswithcode.com/paper/sic-gan-a-self-improving-collaborative-gan
Repo
Framework

Classifier-based Polarity Propagation in a WordNet

Title Classifier-based Polarity Propagation in a WordNet
Authors Jan Koco{'n}, Arkadiusz Janz, Maciej Piasecki
Abstract
Tasks Emotion Recognition, Sentiment Analysis
Published 2018-05-01
URL https://www.aclweb.org/anthology/L18-1665/
PDF https://www.aclweb.org/anthology/L18-1665
PWC https://paperswithcode.com/paper/classifier-based-polarity-propagation-in-a
Repo
Framework

Lightweight Word-Level Confidence Estimation for Neural Interactive Translation Prediction

Title Lightweight Word-Level Confidence Estimation for Neural Interactive Translation Prediction
Authors Rebecca Knowles, Philipp Koehn
Abstract
Tasks Automatic Post-Editing, Machine Translation
Published 2018-03-01
URL https://www.aclweb.org/anthology/W18-2102/
PDF https://www.aclweb.org/anthology/W18-2102
PWC https://paperswithcode.com/paper/lightweight-word-level-confidence-estimation
Repo
Framework

UWB at SemEval-2018 Task 1: Emotion Intensity Detection in Tweets

Title UWB at SemEval-2018 Task 1: Emotion Intensity Detection in Tweets
Authors Pavel P{\v{r}}ib{'a}{\v{n}}, Tom{'a}{\v{s}} Hercig, Ladislav Lenc
Abstract This paper describes our system created for the SemEval-2018 Task 1: Affect in Tweets (AIT-2018). We participated in both the regression and the ordinal classification subtasks for emotion intensity detection in English, Arabic, and Spanish. For the regression subtask we use the AffectiveTweets system with added features using various word embeddings, lexicons, and LDA. For the ordinal classification we additionally use our Brainy system with features using parse tree, POS tags, and morphological features. The most beneficial features apart from word and character n-grams include word embeddings, POS count and morphological features.
Tasks Word Embeddings
Published 2018-06-01
URL https://www.aclweb.org/anthology/S18-1018/
PDF https://www.aclweb.org/anthology/S18-1018
PWC https://paperswithcode.com/paper/uwb-at-semeval-2018-task-1-emotion-intensity
Repo
Framework

Deep JSLC: A Multimodal Corpus Collection for Data-driven Generation of Japanese Sign Language Expressions

Title Deep JSLC: A Multimodal Corpus Collection for Data-driven Generation of Japanese Sign Language Expressions
Authors Heike Brock, Kazuhiro Nakadai
Abstract
Tasks Data Augmentation, Motion Capture
Published 2018-05-01
URL https://www.aclweb.org/anthology/L18-1670/
PDF https://www.aclweb.org/anthology/L18-1670
PWC https://paperswithcode.com/paper/deep-jslc-a-multimodal-corpus-collection-for
Repo
Framework

MapNet: An Allocentric Spatial Memory for Mapping Environments

Title MapNet: An Allocentric Spatial Memory for Mapping Environments
Authors João F. Henriques, Andrea Vedaldi
Abstract Autonomous agents need to reason about the world beyond their instantaneous sensory input. Integrating information over time, however, requires switching from an egocentric representation of a scene to an allocentric one, expressed in the world reference frame. It must also be possible to update the representation dynamically, which requires localizing and registering the sensor with respect to it. In this paper, we develop a differentiable module that satisfies such requirements, while being robust, efficient, and suitable for integration in end-to-end deep networks. The module contains an allocentric spatial memory that can be accessed associatively by feeding to it the current sensory input, resulting in localization, and then updated using an LSTM or similar mechanism. We formulate efficient localization and registration of sensory information as a dual pair of convolution/deconvolution operators in memory space. The map itself is a 2.5D representation of an environment storing information that a deep neural network module learns to distill from RGBD input. The result is a map that contains multi-task information, different from classical approaches to mapping such as structure-from-motion. We present results using synthetic mazes, a dataset of hours of recorded gameplay of the classic game Doom, and the very recent Active Vision Dataset of real images captured from a robot.
Tasks
Published 2018-06-01
URL http://openaccess.thecvf.com/content_cvpr_2018/html/Henriques_MapNet_An_Allocentric_CVPR_2018_paper.html
PDF http://openaccess.thecvf.com/content_cvpr_2018/papers/Henriques_MapNet_An_Allocentric_CVPR_2018_paper.pdf
PWC https://paperswithcode.com/paper/mapnet-an-allocentric-spatial-memory-for
Repo
Framework
comments powered by Disqus