Improved Semantic Representations From Tree-Structured Long Short-Term
Memory Networks
K. Tai, R. Socher, and C. Manning. (2015)cite arxiv:1503.00075Comment: Accepted for publication at ACL 2015.
Abstract
Because of their superior ability to preserve sequence information over time,
Long Short-Term Memory (LSTM) networks, a type of recurrent neural network with
a more complex computational unit, have obtained strong results on a variety of
sequence modeling tasks. The only underlying LSTM structure that has been
explored so far is a linear chain. However, natural language exhibits syntactic
properties that would naturally combine words to phrases. We introduce the
Tree-LSTM, a generalization of LSTMs to tree-structured network topologies.
Tree-LSTMs outperform all existing systems and strong LSTM baselines on two
tasks: predicting the semantic relatedness of two sentences (SemEval 2014, Task
1) and sentiment classification (Stanford Sentiment Treebank).
Description
[1503.00075] Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks
%0 Generic
%1 tai2015improved
%A Tai, Kai Sheng
%A Socher, Richard
%A Manning, Christopher D.
%D 2015
%K lstm network neural recurrent tree
%T Improved Semantic Representations From Tree-Structured Long Short-Term
Memory Networks
%U http://arxiv.org/abs/1503.00075
%X Because of their superior ability to preserve sequence information over time,
Long Short-Term Memory (LSTM) networks, a type of recurrent neural network with
a more complex computational unit, have obtained strong results on a variety of
sequence modeling tasks. The only underlying LSTM structure that has been
explored so far is a linear chain. However, natural language exhibits syntactic
properties that would naturally combine words to phrases. We introduce the
Tree-LSTM, a generalization of LSTMs to tree-structured network topologies.
Tree-LSTMs outperform all existing systems and strong LSTM baselines on two
tasks: predicting the semantic relatedness of two sentences (SemEval 2014, Task
1) and sentiment classification (Stanford Sentiment Treebank).
@misc{tai2015improved,
abstract = {Because of their superior ability to preserve sequence information over time,
Long Short-Term Memory (LSTM) networks, a type of recurrent neural network with
a more complex computational unit, have obtained strong results on a variety of
sequence modeling tasks. The only underlying LSTM structure that has been
explored so far is a linear chain. However, natural language exhibits syntactic
properties that would naturally combine words to phrases. We introduce the
Tree-LSTM, a generalization of LSTMs to tree-structured network topologies.
Tree-LSTMs outperform all existing systems and strong LSTM baselines on two
tasks: predicting the semantic relatedness of two sentences (SemEval 2014, Task
1) and sentiment classification (Stanford Sentiment Treebank).},
added-at = {2018-10-23T10:50:48.000+0200},
author = {Tai, Kai Sheng and Socher, Richard and Manning, Christopher D.},
biburl = {https://www.bibsonomy.org/bibtex/20a5ed49f7420c42d3d9ffe8f0c2afd82/kluger},
description = {[1503.00075] Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks},
interhash = {be0a23242dc9c5b8af6f74672af67310},
intrahash = {0a5ed49f7420c42d3d9ffe8f0c2afd82},
keywords = {lstm network neural recurrent tree},
note = {cite arxiv:1503.00075Comment: Accepted for publication at ACL 2015},
timestamp = {2018-10-23T10:50:48.000+0200},
title = {Improved Semantic Representations From Tree-Structured Long Short-Term
Memory Networks},
url = {http://arxiv.org/abs/1503.00075},
year = 2015
}