Fast(er) Exact Decoding and Global Training for Transition-Based Dependency Parsing via a Minimal Feature Set

Published in

Conference on Empirical Methods in Natural Lang...(2017)

External Links:

Generate Graph DownloadPDF

TL;DR

A minimal feature set for transition-based dependency parsing is presented and the best unlabeled attachment score reported on the Chinese Treebank and the “second-best-in-class” result on the English Penn Treebank are achieved.

Abstract

We first present a minimal feature set for transition-based dependency parsing, continuing a recent trend started by Kiperwasser and Goldberg (2016a) and Cross and Huang (2016a) of using bi-directional LSTM features. We plug our minimal feature set into the dynamic-programming framework of Huang and Sagae (2010) and Kuhlmann et al. (2011) to produce the first implementation of worst-case O(n^3) exact decoders for arc-hybrid and arc-eager transition systems. With our minimal features, we also present O(n^3) global training methods. Finally, using ensembles including our new parsers, we achieve the best unlabeled attachment score reported (to our knowledge) on the Chinese Treebank and the “second-best-in-class” result on the English Penn Treebank.

Authors

Tianze Shi

3 papers

Liang Huang

2 papers

Lillian Lee

1 papers

References66 items

In-Order Transition-based Constituent Parsing

A Minimal Span-Based Neural Constituency Parser

DyNet: The Dynamic Neural Network Toolkit

What Do Recurrent Neural Network Grammars Learn About Syntax?

Deep Biaffine Attention for Neural Dependency Parsing

Fast(er) Exact Decoding and Global Training for Transition-Based Dependency Parsing via a Minimal Feature Set

Published in

Conference on Empirical Methods in Natural Lang...(2017)

External Links:

Generate Graph DownloadPDF

TL;DR

Abstract

Authors

Tianze Shi

3 papers

Liang Huang

2 papers

Lillian Lee

1 papers

References66 items

In-Order Transition-based Constituent Parsing

A Minimal Span-Based Neural Constituency Parser

DyNet: The Dynamic Neural Network Toolkit

What Do Recurrent Neural Network Grammars Learn About Syntax?

Deep Biaffine Attention for Neural Dependency Parsing

Parsing as Language Modeling

Span-Based Constituency Parsing with a Structure-Label System and Provably Optimal Dynamic Oracles

Distilling an Ensemble of Greedy Dependency Parsers into One MST Parser

Bi-directional Attention with Agreement for Dependency Parsing

Graph-based Dependency Parsing with Bidirectional LSTM

Global Neural CCG Parsing with Optimality Guarantees

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

Incremental Parsing with Minimal Features Using Bi-Directional LSTM

Globally Normalized Transition-Based Neural Networks

Simple and Accurate Dependency Parsing Using Bidirectional LSTM Feature Representations

Training with Exploration Improves a Greedy Stack LSTM Parser

Easy-First Dependency Parsing with Hierarchical Tree LSTMs

A Theoretically Grounded Application of Dropout in Recurrent Neural Networks

Optimal Shift-Reduce Constituent Parsing with Structured Perceptron

Structured Training for Neural Network Transition-Based Parsing

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

Adam: A Method for Stochastic Optimization

GloVe: Global Vectors for Word Representation

The Inside-Outside Recursive Neural Network model for Dependency Parsing

Training Deterministic Parsers with Non-Deterministic Oracles

Distributed Representations of Words and Phrases and their Compositionality

Optimal Incremental Parsing via Best-First Dynamic Programming

Squibs: Going to the Roots of Dependency Parsing

A Dynamic Oracle for Arc-Eager Dependency Parsing

Dependency Parsing Schemata and Mildly Non-Projective Dependency Parsing

Dynamic Programming Algorithms for Transition-Based Dependency Parsers

Syntactic Processing Using the Generalized Perceptron and Beam Search

Dynamic Programming for Linear-Time Incremental Parsing

Understanding the difficulty of training deep feedforward neural networks

Algorithms for Deterministic Incremental Dependency Parsing

A Tale of Two Parsers: Investigating and Combining Graph-based and Transition-based Dependency Parsing

A Deductive Approach to Dependency Parsing

Dependency Parsing and Domain Adaptation with LR Models and Parser Ensembles

Experiments with a Higher-Order Projective Dependency Parser

A Best-First Probabilistic Shift-Reduce Parser

Online Learning of Approximate Dependency Parsing Algorithms

Learning structured prediction models: a large margin approach

Incrementality in Deterministic Dependency Parsing

Discriminative Training of a Neural Network Statistical Parser

Accurate Unlexicalized Parsing

Inducing History Representations for Broad Coverage Statistical Parsing

Feature-Rich Part-of-Speech Tagging with a Cyclic Dependency Network

Statistical Dependency Analysis with Support Vector Machines

Building a Large-Scale Annotated Chinese Corpus

Efficient Parsing for Bilexical Context-Free Grammars and Head Automaton Grammars

Three New Probabilistic Models for Dependency Parsing: An Exploration

Principles and Implementation of Deductive Parsing

Building a Large Annotated Corpus of English: The Penn Treebank

Parsing as Deduction

Recognition and Parsing of Context-Free Languages in Time n^3

An Efficient Recognition and Syntax-Analysis Algorithm for Context-Free Languages

Combining Global Models for Parsing Universal Dependencies

Dropout: a simple way to prevent neural networks from overfitting

A Fast and Accurate Dependency Parser using Neural Networks

Synthesis Lectures on Human Language Technologies

Linguistic Structure Prediction

Stanford typed dependencies manual

Dependency Parsing

Framewise phoneme classification with bidirectional LSTM and other neural network architectures

An Efficient Algorithm for Projective Dependency Parsing

Programming languages and their compilers: Preliminary notes

Field of Study

Computer Science

Journal Information

Name

ArXiv

Volume

abs/2005.00687

Venue Information

Name

Conference on Empirical Methods in Natural Language Processing

Type

conference

URL

https://www.aclweb.org/portal/emnlp

Alternate Names

Empir Method Nat Lang Process
Empirical Methods in Natural Language Processing
Conf Empir Method Nat Lang Process
EMNLP