The celebrated Sequence to Sequence learning (Seq2Seq) technique and
its numerous variants achieve excellent performance on many tasks. However,
many machine learning tasks have inputs naturally represented as graphs;
existing Seq2Seq models face a significant challenge in achieving accurate
conversion from graph form to the appropriate sequence. To address this
challenge, we introduce a general end-to-end graph-to-sequence neural
encoder-decoder architecture that maps an input graph to a sequence of vectors
and uses an attention-based LSTM method to decode the target sequence from
these vectors. Our method first generates the node and graph embeddings using
an improved graph-based neural network with a novel aggregation strategy to
incorporate edge direction information in the node embeddings. We further
introduce an attention mechanism that aligns node embeddings and the decoding
sequence to better cope with large graphs. Experimental results on bAbI,
Shortest Path, and Natural Language Generation tasks demonstrate that our model
achieves state-of-the-art performance and significantly outperforms baseline
systems; using the proposed aggregation strategy, the model can converge
rapidly to the optimal performance.
Description
[1804.00823] Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks
%0 Generic
%1 xu2018graph2seq
%A Xu, Kun
%A Wu, Lingfei
%A Wang, Zhiguo
%A Feng, Yansong
%A Witbrock, Michael
%A Sheinin, Vadim
%D 2018
%K attention graph journey network networks neural prediction
%T Graph2Seq: Graph to Sequence Learning with Attention-based Neural
Networks
%U http://arxiv.org/abs/1804.00823
%X The celebrated Sequence to Sequence learning (Seq2Seq) technique and
its numerous variants achieve excellent performance on many tasks. However,
many machine learning tasks have inputs naturally represented as graphs;
existing Seq2Seq models face a significant challenge in achieving accurate
conversion from graph form to the appropriate sequence. To address this
challenge, we introduce a general end-to-end graph-to-sequence neural
encoder-decoder architecture that maps an input graph to a sequence of vectors
and uses an attention-based LSTM method to decode the target sequence from
these vectors. Our method first generates the node and graph embeddings using
an improved graph-based neural network with a novel aggregation strategy to
incorporate edge direction information in the node embeddings. We further
introduce an attention mechanism that aligns node embeddings and the decoding
sequence to better cope with large graphs. Experimental results on bAbI,
Shortest Path, and Natural Language Generation tasks demonstrate that our model
achieves state-of-the-art performance and significantly outperforms baseline
systems; using the proposed aggregation strategy, the model can converge
rapidly to the optimal performance.
@misc{xu2018graph2seq,
abstract = {The celebrated \emph{Sequence to Sequence learning (Seq2Seq)} technique and
its numerous variants achieve excellent performance on many tasks. However,
many machine learning tasks have inputs naturally represented as graphs;
existing Seq2Seq models face a significant challenge in achieving accurate
conversion from graph form to the appropriate sequence. To address this
challenge, we introduce a general end-to-end graph-to-sequence neural
encoder-decoder architecture that maps an input graph to a sequence of vectors
and uses an attention-based LSTM method to decode the target sequence from
these vectors. Our method first generates the node and graph embeddings using
an improved graph-based neural network with a novel aggregation strategy to
incorporate edge direction information in the node embeddings. We further
introduce an attention mechanism that aligns node embeddings and the decoding
sequence to better cope with large graphs. Experimental results on bAbI,
Shortest Path, and Natural Language Generation tasks demonstrate that our model
achieves state-of-the-art performance and significantly outperforms baseline
systems; using the proposed aggregation strategy, the model can converge
rapidly to the optimal performance.},
added-at = {2018-10-29T10:01:21.000+0100},
author = {Xu, Kun and Wu, Lingfei and Wang, Zhiguo and Feng, Yansong and Witbrock, Michael and Sheinin, Vadim},
biburl = {https://www.bibsonomy.org/bibtex/20b5d0bc24bd82e1b20a973c140c02632/robax},
description = {[1804.00823] Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks},
interhash = {8901c24183714b330aedbacd4069373a},
intrahash = {0b5d0bc24bd82e1b20a973c140c02632},
keywords = {attention graph journey network networks neural prediction},
note = {cite arxiv:1804.00823Comment: 16 pages, 3 figures, 4 tables},
timestamp = {2018-10-29T10:01:21.000+0100},
title = {Graph2Seq: Graph to Sequence Learning with Attention-based Neural
Networks},
url = {http://arxiv.org/abs/1804.00823},
year = 2018
}