Globally Normalized Transition-Based Neural Networks (2016-03-19T00:00:00.000000Z)

TL;DR

A globally normalized transition-based neural network model that achieves state-of-the-art part- of-speech tagging, dependency parsing and sentence compression results is introduced.

Abstract

We introduce a globally normalized transition-based neural network model that achieves state-of-the-art part-of-speech tagging, dependency parsing and sentence compression results. Our model is a simple feed-forward neural network that operates on a task-specific transition system, yet achieves comparable or better accuracies than recurrent models. We discuss the importance of global as opposed to local normalization: a key insight is that the label bias problem implies that globally normalized models can be strictly more expressive than locally normalized models.

Authors

Chris Alberti

8 papers

Michael Collins

7 papers

Slav Petrov

5 papers

TL;DR

Abstract

Authors

References52 items

Efficient Structured Inference for Transition-Based Parsing with Neural Networks and Error States

Improved Transition-Based Parsing and Tagging with Neural Networks

Sentence Compression by Deletion with LSTMs

Finding Function in Form: Compositional Character Models for Open Vocabulary Word Representation

Bidirectional LSTM-CRF Models for Sequence Tagging

Improved Transition-based Parsing by Modeling Characters instead of Words with LSTMs

End-to-end learning of semantic role labeling using recurrent neural networks

Neural CRF Parsing

A Neural Probabilistic Structured-Prediction Model for Transition-Based Dependency Parsing

Transition-based Neural Constituent Parsing

Incremental Recurrent Neural Network Dependency Parser with Search-based Discriminative Training

Structured Training for Neural Network Transition-Based Parsing

Transition-Based Dependency Parsing with Stack Long Short-Term Memory

Conditional Random Fields as Recurrent Neural Networks

Grammar as a Foreign Language

Low-Rank Tensors for Scoring Dependency Structures

Enforcing Structural Diversity in Cube-pruned Dependency Parsing

Recurrent conditional random field for language understanding

Training Deterministic Parsers with Non-Deterministic Oracles

Turning on the Turbo: Fast Third-Order Non-Projective Turbo Parsers

A Transition-Based System for Joint Part-of-Speech Tagging and Labeled Non-Projective Dependency Parsing

Natural Language Processing (Almost) from Scratch

No-Regret Reductions for Imitation Learning and Structured Prediction

Neural conditional random fields

Conditional Neural Fields

Non-Projective Dependency Parsing in Expected Linear Time

The CoNLL-2009 Shared Task: Syntactic and Semantic Dependencies in Multiple Languages

Search-based structured prediction

Structure compilation: trading structure for features

Weighted and Probabilistic Context-Free Grammars Are Equally Expressive

QuestionBank: Creating a Corpus of Parse-Annotated Questions

OntoNotes: The 90% Solution

Generating Typed Dependency Parses from Phrase Structure Parses

Discriminative Training of a Neural Network Statistical Parser

Incremental Parsing with the Perceptron Algorithm

Head-Driven Statistical Models for Natural Language Parsing

Inducing History Representations for Broad Coverage Statistical Parsing

Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data

Relating Probabilistic Grammars and Automata

Statistical Properties of Probabilistic Context-Free Grammars

Long Short-Term Memory

Global training of document processing systems using graph transformer networks

Building a Large Annotated Corpus of English: The Penn Treebank

CONDITIONAL RANDOM FIELDS

Colmenares, Łukasz Kaiser, and Oriol Vinyals

A Fast and Accurate Dependency Parser using Neural Networks

Overview of the 2012 Shared Task on Parsing the Web

Inductive Dependency Parsing

Graph transformer networks for image recognition

Gradient-based learning applied to document recognition

Une approche th´eorique de lap-prentissage connexionniste: Applications `a la reconnaissance de la parole

Conference on Empirical Methods

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names