October 15, 2019

2579 words 13 mins read

Paper Group NANR 266

Paper Group NANR 266

State Abstractions for Lifelong Reinforcement Learning. Learning to select examples for program synthesis. Learning non-linear transform with discriminative and minimum information loss priors. Corpus Phonetics: Past, Present, and Future. Visualization of the occurrence trend of infectious diseases using Twitter. Trust Your Model: Light Field Depth …

State Abstractions for Lifelong Reinforcement Learning

Title State Abstractions for Lifelong Reinforcement Learning
Authors David Abel, Dilip Arumugam, Lucas Lehnert, Michael Littman
Abstract In lifelong reinforcement learning, agents must effectively transfer knowledge across tasks while simultaneously addressing exploration, credit assignment, and generalization. State abstraction can help overcome these hurdles by compressing the representation used by an agent, thereby reducing the computational and statistical burdens of learning. To this end, we here develop theory to compute and use state abstractions in lifelong reinforcement learning. We introduce two new classes of abstractions: (1) transitive state abstractions, whose optimal form can be computed efficiently, and (2) PAC state abstractions, which are guaranteed to hold with respect to a distribution of tasks. We show that the joint family of transitive PAC abstractions can be acquired efficiently, preserve near optimal-behavior, and experimentally reduce sample complexity in simple domains, thereby yielding a family of desirable abstractions for use in lifelong reinforcement learning. Along with these positive results, we show that there are pathological cases where state abstractions can negatively impact performance.
Tasks
Published 2018-07-01
URL https://icml.cc/Conferences/2018/Schedule?showEvent=2087
PDF http://proceedings.mlr.press/v80/abel18a/abel18a.pdf
PWC https://paperswithcode.com/paper/state-abstractions-for-lifelong-reinforcement
Repo
Framework

Learning to select examples for program synthesis

Title Learning to select examples for program synthesis
Authors Yewen Pu, Zachery Miranda, Armando Solar-Lezama, Leslie Pack Kaelbling
Abstract Program synthesis is a class of regression problems where one seeks a solution, in the form of a source-code program, that maps the inputs to their corresponding outputs exactly. Due to its precise and combinatorial nature, it is commonly formulated as a constraint satisfaction problem, where input-output examples are expressed constraints, and solved with a constraint solver. A key challenge of this formulation is that of scalability: While constraint solvers work well with few well-chosen examples, constraining the entire set of example constitutes a significant overhead in both time and memory. In this paper we address this challenge by constructing a representative subset of examples that is both small and is able to constrain the solver sufficiently. We build the subset one example at a time, using a trained discriminator to predict the probability of unchosen input-output examples conditioned on the chosen input-output examples, adding the least probable example to the subset. Experiment on a diagram drawing domain shows our approach produces subset of examples that are small and representative for the constraint solver.
Tasks Program Synthesis
Published 2018-01-01
URL https://openreview.net/forum?id=B1CQGfZ0b
PDF https://openreview.net/pdf?id=B1CQGfZ0b
PWC https://paperswithcode.com/paper/learning-to-select-examples-for-program
Repo
Framework

Learning non-linear transform with discriminative and minimum information loss priors

Title Learning non-linear transform with discriminative and minimum information loss priors
Authors Dimche Kostadinov, Slava Voloshynovskiy
Abstract This paper proposes a novel approach for learning discriminative and sparse representations. It consists of utilizing two different models. A predefined number of non-linear transform models are used in the learning stage, and one sparsifying transform model is used at test time. The non-linear transform models have discriminative and minimum information loss priors. A novel measure related to the discriminative prior is proposed and defined on the support intersection for the transform representations. The minimum information loss prior is expressed as a constraint on the conditioning and the expected coherence of the transform matrix. An equivalence between the non-linear models and the sparsifying model is shown only when the measure that is used to define the discriminative prior goes to zero. An approximation of the measure used in the discriminative prior is addressed, connecting it to a similarity concentration. To quantify the discriminative properties of the transform representation, we introduce another measure and present its bounds. Reflecting the discriminative quality of the transform representation we name it as discrimination power. To support and validate the theoretical analysis a practical learning algorithm is presented. We evaluate the advantages and the potential of the proposed algorithm by a computer simulation. A favorable performance is shown considering the execution time, the quality of the representation, measured by the discrimination power and the recognition accuracy in comparison with the state-of-the-art methods of the same category.
Tasks
Published 2018-01-01
URL https://openreview.net/forum?id=SJzmJEq6W
PDF https://openreview.net/pdf?id=SJzmJEq6W
PWC https://paperswithcode.com/paper/learning-non-linear-transform-with
Repo
Framework

Corpus Phonetics: Past, Present, and Future

Title Corpus Phonetics: Past, Present, and Future
Authors Mark Liberman
Abstract Invited talk
Tasks
Published 2018-08-01
URL https://www.aclweb.org/anthology/W18-3801/
PDF https://www.aclweb.org/anthology/W18-3801
PWC https://paperswithcode.com/paper/corpus-phonetics-past-present-and-future
Repo
Framework

Visualization of the occurrence trend of infectious diseases using Twitter

Title Visualization of the occurrence trend of infectious diseases using Twitter
Authors Ryusei Matsumoto, Minoru Yoshida, Kazuyuki Matsumoto, Hironobu Matsuda, Kenji Kita
Abstract
Tasks
Published 2018-05-01
URL https://www.aclweb.org/anthology/L18-1081/
PDF https://www.aclweb.org/anthology/L18-1081
PWC https://paperswithcode.com/paper/visualization-of-the-occurrence-trend-of
Repo
Framework

Trust Your Model: Light Field Depth Estimation With Inline Occlusion Handling

Title Trust Your Model: Light Field Depth Estimation With Inline Occlusion Handling
Authors Hendrik Schilling, Maximilian Diebold, Carsten Rother, Bernd Jähne
Abstract We address the problem of depth estimation from light-field images. Our main contribution is a new way to handle occlusions which improves general accuracy and quality of object borders. In contrast to all prior work we work with a model which directly incorporates both depth and occlusion, using a local optimization scheme based on the PatchMatch algorithm. The key benefit of this joint approach is that we utilize all available data, and not erroneously discard valuable information in pre-processing steps. We see the benefit of our approach not only at improved object boundaries, but also at smooth surface reconstruction, where we outperform even methods which focus on good surface regularization. We have evaluated our method on a public light-field dataset, where we achieve state-of-the-art results in nine out of twelve error metrics, with a close tie for the remaining three.
Tasks Depth Estimation
Published 2018-06-01
URL http://openaccess.thecvf.com/content_cvpr_2018/html/Schilling_Trust_Your_Model_CVPR_2018_paper.html
PDF http://openaccess.thecvf.com/content_cvpr_2018/papers/Schilling_Trust_Your_Model_CVPR_2018_paper.pdf
PWC https://paperswithcode.com/paper/trust-your-model-light-field-depth-estimation
Repo
Framework

Linearly Constrained Weights: Resolving the Vanishing Gradient Problem by Reducing Angle Bias

Title Linearly Constrained Weights: Resolving the Vanishing Gradient Problem by Reducing Angle Bias
Authors Takuro Kutsuna
Abstract In this paper, we first identify \textit{angle bias}, a simple but remarkable phenomenon that causes the vanishing gradient problem in a multilayer perceptron (MLP) with sigmoid activation functions. We then propose \textit{linearly constrained weights (LCW)} to reduce the angle bias in a neural network, so as to train the network under the constraints that the sum of the elements of each weight vector is zero. A reparameterization technique is presented to efficiently train a model with LCW by embedding the constraints on weight vectors into the structure of the network. Interestingly, batch normalization (Ioffe & Szegedy, 2015) can be viewed as a mechanism to correct angle bias. Preliminary experiments show that LCW helps train a 100-layered MLP more efficiently than does batch normalization.
Tasks
Published 2018-01-01
URL https://openreview.net/forum?id=HylgYB3pZ
PDF https://openreview.net/pdf?id=HylgYB3pZ
PWC https://paperswithcode.com/paper/linearly-constrained-weights-resolving-the
Repo
Framework

Authorship Attribution By Consensus Among Multiple Features

Title Authorship Attribution By Consensus Among Multiple Features
Authors Jagadeesh Patchala, Raj Bhatnagar
Abstract Most existing research on authorship attribution uses various lexical, syntactic and semantic features. In this paper we demonstrate an effective template-based approach for combining various syntactic features of a document for authorship analysis. The parse-tree based features that we propose are independent of the topic of a document and reflect the innate writing styles of authors. We show that the use of templates including sub-trees of parse trees in conjunction with other syntactic features result in improved author attribution rates. Another contribution is the demonstration that Dempster{'}s rule based combination of evidence from syntactic features performs better than other evidence-combination methods. We also demonstrate that our methodology works well for the case where actual author is not included in the candidate author set.
Tasks
Published 2018-08-01
URL https://www.aclweb.org/anthology/C18-1234/
PDF https://www.aclweb.org/anthology/C18-1234
PWC https://paperswithcode.com/paper/authorship-attribution-by-consensus-among
Repo
Framework

USI-IR at IEST 2018: Sequence Modeling and Pseudo-Relevance Feedback for Implicit Emotion Detection

Title USI-IR at IEST 2018: Sequence Modeling and Pseudo-Relevance Feedback for Implicit Emotion Detection
Authors Esteban R{'\i}ssola, Anastasia Giachanou, Fabio Crestani
Abstract This paper describes the participation of USI-IR in WASSA 2018 Implicit Emotion Shared Task. We propose a relevance feedback approach employing a sequential model (biLSTM) and word embeddings derived from a large collection of tweets. To this end, we assume that the top-\textit{k} predictions produce at a first classification step are correct (based on the model accuracy) and use them as new examples to re-train the network.
Tasks Emotion Recognition, Sentiment Analysis, Word Embeddings
Published 2018-10-01
URL https://www.aclweb.org/anthology/W18-6233/
PDF https://www.aclweb.org/anthology/W18-6233
PWC https://paperswithcode.com/paper/usi-ir-at-iest-2018-sequence-modeling-and
Repo
Framework

Where Have I Heard This Story Before? Identifying Narrative Similarity in Movie Remakes

Title Where Have I Heard This Story Before? Identifying Narrative Similarity in Movie Remakes
Authors Snigdha Chaturvedi, Shashank Srivastava, Dan Roth
Abstract People can identify correspondences between narratives in everyday life. For example, an analogy with the Cinderella story may be made in describing the unexpected success of an underdog in seemingly different stories. We present a new task and dataset for story understanding: identifying instances of similar narratives from a collection of narrative texts. We present an initial approach for this problem, which finds correspondences between narratives in terms of plot events, and resemblances between characters and their social relationships. Our approach yields an 8{%} absolute improvement in performance over a competitive information-retrieval baseline on a novel dataset of plot summaries of 577 movie remakes from Wikipedia.
Tasks Information Retrieval
Published 2018-06-01
URL https://www.aclweb.org/anthology/N18-2106/
PDF https://www.aclweb.org/anthology/N18-2106
PWC https://paperswithcode.com/paper/where-have-i-heard-this-story-before
Repo
Framework

Convolutional Interaction Network for Natural Language Inference

Title Convolutional Interaction Network for Natural Language Inference
Authors Jingjing Gong, Xipeng Qiu, Xinchi Chen, Dong Liang, Xuanjing Huang
Abstract Attention-based neural models have achieved great success in natural language inference (NLI). In this paper, we propose the Convolutional Interaction Network (CIN), a general model to capture the interaction between two sentences, which can be an alternative to the attention mechanism for NLI. Specifically, CIN encodes one sentence with the filters dynamically generated based on another sentence. Since the filters may be designed to have various numbers and sizes, CIN can capture more complicated interaction patterns. Experiments on three large datasets demonstrate CIN{'}s efficacy.
Tasks Information Retrieval, Natural Language Inference, Question Answering
Published 2018-10-01
URL https://www.aclweb.org/anthology/D18-1186/
PDF https://www.aclweb.org/anthology/D18-1186
PWC https://paperswithcode.com/paper/convolutional-interaction-network-for-natural
Repo
Framework

DisNet: A novel method for distance estimation from monocular camera

Title DisNet: A novel method for distance estimation from monocular camera
Authors Muhammad Abdul Haseeb, Jianyu Guan, Danijela Ristić-Durrant, Axel Gräser
Abstract In this paper, a machine learning setup that provides the obstacle detection system with a method to estimate the distance from the monocular camera to the object viewed with the camera is presented. In particular, the preliminary results of on-going research to allow the onboard multisensory system, which is under development within H2020 Shift2Rail project SMART, to autonomously learn distances to objects, possible obstacles on the rail tracks ahead of the locomotive are given. The presented distance estimation system is based on Multi Hidden-Layer Neural Network, named DisNet, which is used to learn and predict the distance between the object and the camera sensor. The DisNet was trained using a supervised learning technique where the input features were manually calculated parameters of the object bounding boxes resulted from the YOLO object classifier and outputs were the accurate 3D laser scanner measurements of the distances to objects in the recorded scene. The presented DisNet-based distance estimation system was evaluated on the images of railway scenes as well as on the images of a road scene. Shown results demonstrate a general nature of the proposed DisNet system that enables its use for the estimation of distances to objects imaged with different types of monocular cameras.
Tasks Depth Estimation
Published 2018-10-01
URL https://webcache.googleusercontent.com/search?q=cache:x7y3KzAxZdgJ:https://project.inria.fr/ppniv18/files/2018/10/paper22.pdf+&cd=2&hl=en&ct=clnk&gl=ca
PDF https://project.inria.fr/ppniv18/files/2018/10/paper22.pdf
PWC https://paperswithcode.com/paper/disnet-a-novel-method-for-distance-estimation
Repo
Framework

Propagating LSTM: 3D Pose Estimation based on Joint Interdependency

Title Propagating LSTM: 3D Pose Estimation based on Joint Interdependency
Authors Kyoungoh Lee, Inwoong Lee, Sanghoon Lee
Abstract We present a novel 3D pose estimation method based on joint interdependency (JI) for acquiring 3D joints from the human pose of an RGB image. The JI incorporates the body part based structural connectivity of joints to learn the high spatial correlation of human posture on our method. Towards this goal, we propose a new long short-term memory (LSTM)-based deep learning architecture named propagating LSTM networks (p-LSTMs), where each LSTM is connected sequentially to reconstruct 3D depth from the centroid to edge joints through learning the intrinsic JI. In the first LSTM, the seed joints of 3D pose are created and reconstructed into the whole-body joints through the connected LSTMs. Utilizing the p-LSTMs, we achieve the higher accuracy of about 11.2% than state-of-the-art methods on the largest publicly available database. Importantly, we demonstrate that the JI drastically reduces the structural errors at body edges, thereby leads to a significant improvement.
Tasks 3D Human Pose Estimation, 3D Pose Estimation, Pose Estimation
Published 2018-09-01
URL http://openaccess.thecvf.com/content_ECCV_2018/html/Kyoungoh_Lee_Propagating_LSTM_3D_ECCV_2018_paper.html
PDF http://openaccess.thecvf.com/content_ECCV_2018/papers/Kyoungoh_Lee_Propagating_LSTM_3D_ECCV_2018_paper.pdf
PWC https://paperswithcode.com/paper/propagating-lstm-3d-pose-estimation-based-on
Repo
Framework

Learning to Generate Word Representations using Subword Information

Title Learning to Generate Word Representations using Subword Information
Authors Yeachan Kim, Kang-Min Kim, Ji-Min Lee, SangKeun Lee
Abstract Distributed representations of words play a major role in the field of natural language processing by encoding semantic and syntactic information of words. However, most existing works on learning word representations typically regard words as individual atomic units and thus are blind to subword information in words. This further gives rise to a difficulty in representing out-of-vocabulary (OOV) words. In this paper, we present a character-based word representation approach to deal with this limitation. The proposed model learns to generate word representations from characters. In our model, we employ a convolutional neural network and a highway network over characters to extract salient features effectively. Unlike previous models that learn word representations from a large corpus, we take a set of pre-trained word embeddings and generalize it to word entries, including OOV words. To demonstrate the efficacy of the proposed model, we perform both an intrinsic and an extrinsic task which are word similarity and language modeling, respectively. Experimental results show clearly that the proposed model significantly outperforms strong baseline models that regard words or their subwords as atomic units. For example, we achieve as much as 18.5{%} improvement on average in perplexity for morphologically rich languages compared to strong baselines in the language modeling task.
Tasks Chunking, Language Modelling, Named Entity Recognition, Question Answering, Text Classification, Word Embeddings
Published 2018-08-01
URL https://www.aclweb.org/anthology/C18-1216/
PDF https://www.aclweb.org/anthology/C18-1216
PWC https://paperswithcode.com/paper/learning-to-generate-word-representations
Repo
Framework

Image Quality Assessment Techniques Improve Training and Evaluation of Energy-Based Generative Adversarial Networks

Title Image Quality Assessment Techniques Improve Training and Evaluation of Energy-Based Generative Adversarial Networks
Authors Michael O. Vertolli, Jim Davies
Abstract We propose a new, multi-component energy function for energy-based Generative Adversarial Networks (GANs) based on methods from the image quality assessment literature. Our approach expands on the Boundary Equilibrium Generative Adversarial Network (BEGAN) by outlining some of the short-comings of the original energy and loss functions. We address these short-comings by incorporating an l1 score, the Gradient Magnitude Similarity score, and a chrominance score into the new energy function. We then provide a set of systematic experiments that explore its hyper-parameters. We show that each of the energy function’s components is able to represent a slightly different set of features, which require their own evaluation criteria to assess whether they have been adequately learned. We show that models using the new energy function are able to produce better image representations than the BEGAN model in predicted ways.
Tasks Image Quality Assessment
Published 2018-01-01
URL https://openreview.net/forum?id=ryzm6BATZ
PDF https://openreview.net/pdf?id=ryzm6BATZ
PWC https://paperswithcode.com/paper/image-quality-assessment-techniques-improve
Repo
Framework
comments powered by Disqus