ReasoNet: Learning to Stop Reading in Machine Comprehension

Abstract

Teaching a computer to read and answer general questions pertaining to a document is a challenging yet unsolved problem. In this paper, we describe a novel neural network architecture called the Reasoning Network (ReasoNet) for machine comprehension tasks. ReasoNets make use of multiple turns to effectively exploit and then reason over the relation among queries, documents, and answers. Different from previous approaches using a fixed number of turns during inference, ReasoNets introduce a termination state to relax this constraint on the reasoning depth. With the use of reinforcement learning, ReasoNets can dynamically determine whether to continue the comprehension process after digesting intermediate results, or to terminate reading when it concludes that existing information is adequate to produce an answer. ReasoNets have achieved exceptional performance in machine comprehension datasets, including unstructured CNN and Daily Mail datasets, the Stanford SQuAD dataset, and a structured Graph Reachability dataset.

in KDD 2017

References in corpus (12)

• ADADELTA: An Adaptive Learning Rate Method
• Bidirectional Attention Flow for Machine Comprehension
• Recurrent Models of Visual Attention
• Dynamic Coattention Networks For Question Answering
• Machine Comprehension Using Match-LSTM and Answer Pointer
• Learning Recurrent Span Representations for Extractive Question Answering
• A Thorough Examination of the CNN/Daily Mail Reading Comprehension Task
• Multi-Perspective Context Matching for Machine Comprehension
• Words or Characters? Fine-grained Gating for Reading Comprehension
• Exploring Question Understanding and Adaptation in Neural-Network-Based Question Answering
• Iterative Alternating Neural Attention for Machine Reading
• Separating Answers from Queries for Neural Reading Comprehension

Cited by in corpus (89)

• Recent Trends in Deep Learning Based Natural Language Processing
• Attention in Natural Language Processing
• Deep Reinforcement Learning: An Overview
• QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension
• Go for a Walk and Arrive at the Answer: Reasoning Over Paths in Knowledge Bases using Reinforcement Learning
• Deep Learning Based Text Classification: A Comprehensive Review
• Adversarial Examples for Evaluating Reading Comprehension Systems
• A Text Classification Framework for Simple and Effective Early Depression Detection Over Social Media Streams
• End-to-End Differentiable Proving
• Multi-Perspective Context Matching for Machine Comprehension
• An Interpretable Reasoning Network for Multi-Relation Question Answering
• DCN+: Mixed Objective and Deep Residual Coattention for Question Answering
• Exploring Graph-structured Passage Representation for Multi-hop Reading Comprehension with Graph Neural Networks
• A Joint Model for Question Answering and Question Generation
• Focal Visual-Text Attention for Visual Question Answering
• MEMEN: Multi-layer Embedding with Memory Networks for Machine Comprehension
• Multi-hop Question Answering via Reasoning Chains
• Alchemy: A Quantum Chemistry Dataset for Benchmarking AI Models
• A Unified MRC Framework for Named Entity Recognition
• A Unified Query-based Generative Model for Question Generation and Question Answering
• Entity-Relation Extraction as Multi-Turn Question Answering
• Machine Reading Comprehension: The Role of Contextualized Language Models and Beyond
• Making Neural QA as Simple as Possible but not Simpler
• Reinforced Mnemonic Reader for Machine Reading Comprehension
• Dynamic Neural Networks: A Survey
• U-Net: Machine Reading Comprehension with Unanswerable Questions
• Stochastic Answer Networks for Machine Reading Comprehension
• Multi-range Reasoning for Machine Comprehension
• Dice Loss for Data-imbalanced NLP Tasks
• Description Based Text Classification with Reinforcement Learning
• A Comparative Study of Word Embeddings for Reading Comprehension
• Neural Machine Reading Comprehension: Methods and Trends
• Machine Reading Comprehension: a Literature Review
• Cognitive Graph for Multi-Hop Reading Comprehension at Scale
• Efficient and Robust Question Answering from Minimal Context over Documents
• Multi-step Reasoning via Recurrent Dual Attention for Visual Dialog
• Reinforced Self-Attention Network: a Hybrid of Hard and Soft Attention for Sequence Modeling
• Focused Hierarchical RNNs for Conditional Sequence Processing
• Learning Visually-Grounded Semantics from Contrastive Adversarial Samples
• MemexQA: Visual Memex Question Answering
• A Survey on Neural Machine Reading Comprehension
• The Effect of Natural Distribution Shift on Question Answering Models
• Unstructured Text Enhanced Open-domain Dialogue System: A Systematic Survey
• Dynamic Fusion Networks for Machine Reading Comprehension
• Comparison of State-of-the-Art Deep Learning APIs for Image Multi-Label Classification using Semantic Metrics
• Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index
• Smarnet: Teaching Machines to Read and Comprehend Like Human
• Asking questions on handwritten document collections
• Multi-Passage Machine Reading Comprehension with Cross-Passage Answer Verification
• Interpretable Multi-Step Reasoning with Knowledge Extraction on Complex Healthcare Question Answering
• Open Domain Question Answering Using Early Fusion of Knowledge Bases and Text
• Joint Training of Candidate Extraction and Answer Selection for Reading Comprehension
• A Survey of Document Grounded Dialogue Systems (DGDS)
• Explicit Utilization of General Knowledge in Machine Reading Comprehension
• Biomedical named entity recognition using BERT in the machine reading comprehension framework
• Understanding Dataset Design Choices for Multi-hop Reasoning
• Do Multi-hop Readers Dream of Reasoning Chains?
• Link Prediction using Embedded Knowledge Graphs
• MEMO: A Deep Network for Flexible Combination of Episodic Memories
• SRQA: Synthetic Reader for Factoid Question Answering
• QAInfomax: Learning Robust Question Answering System by Mutual Information Maximization
• Adversarial Domain Adaptation for Machine Reading Comprehension
• Multi-task Learning with Sample Re-weighting for Machine Reading Comprehension
• Making Neural Machine Reading Comprehension Faster
• Leap-LSTM: Enhancing Long Short-Term Memory for Text Categorization
• Large-scale Multi-granular Concept Extraction Based on Machine Reading Comprehension
• Neural Skill Transfer from Supervised Language Tasks to Reading Comprehension
• Document-editing Assistants and Model-based Reinforcement Learning as a Path to Conversational AI
• Learning to Search in Long Documents Using Document Structure
• Interactive Semantic Parsing for If-Then Recipes via Hierarchical Reinforcement Learning
• Deep Neural Networks for Relation Extraction
• Robust and Scalable Differentiable Neural Computer for Question Answering
• Integrated Triaging for Fast Reading Comprehension
• Attentive Convolution: Equipping CNNs with RNN-style Attention Mechanisms
• Tackling Graphical NLP problems with Graph Recurrent Networks
• Keyword-based Query Comprehending via Multiple Optimized-Demand Augmentation
• An Empirical Study of Content Understanding in Conversational Question Answering
• Multi-hop Reading Comprehension across Documents with Path-based Graph Convolutional Network
• A Study of the Tasks and Models in Machine Reading Comprehension
• Interactive Machine Comprehension with Information Seeking Agents
• Query-Based Named Entity Recognition
• BiPaR: A Bilingual Parallel Dataset for Multilingual and Cross-lingual Reading Comprehension on Novels
• An Empirical Analysis of Multiple-Turn Reasoning Strategies in Reading Comprehension Tasks
• Propagate-Selector: Detecting Supporting Sentences for Question Answering via Graph Neural Networks
• Automatic Task Requirements Writing Evaluation via Machine Reading Comprehension
• CAESAR: Context Awareness Enabled Summary-Attentive Reader
• A Survey on Green Deep Learning
• A3Net: Adversarial-and-Attention Network for Machine Reading Comprehension
• Incremental Reading for Question Answering