Top-down Tree Long Short-Term Memory Networks

October 31, 2015 · Entered Twilight · 🏛 North American Chapter of the Association for Computational Linguistics

"Last commit was 9.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: LICENSE, OLD_VERSION.md, README.md, conllx_scripts, dataset, dep_rerank.lua, experiments, gpu_lock.py, init.lua, layers, main.lua, main_bid.lua, main_nce.lua, msr_scripts, nnets, rerank.lua, sampler.lua, scripts, train_mlp.lua, utils

Authors Xingxing Zhang, Liang Lu, Mirella Lapata arXiv ID 1511.00060 Category cs.CL: Computation & Language Cross-listed cs.LG Citations 101 Venue North American Chapter of the Association for Computational Linguistics Repository https://github.com/XingxingZhang/td-treelstm ⭐ 83 Last Checked 1 month ago

Abstract

Long Short-Term Memory (LSTM) networks, a type of recurrent neural network with a more complex computational unit, have been successfully applied to a variety of sequence modeling tasks. In this paper we develop Tree Long Short-Term Memory (TreeLSTM), a neural network model based on LSTM, which is designed to predict a tree rather than a linear sequence. TreeLSTM defines the probability of a sentence by estimating the generation probability of its dependency tree. At each time step, a node is generated based on the representation of the generated sub-tree. We further enhance the modeling power of TreeLSTM by explicitly representing the correlations between left and right dependents. Application of our model to the MSR sentence completion challenge achieves results beyond the current state of the art. We also report results on dependency parsing reranking achieving competitive performance.