Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks

Abstract

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 novel general end-to-end graph-to-sequence neural encoder-decoder model 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 existing graph neural networks, Seq2Seq, and Tree2Seq models; using the proposed bi-directional node embedding aggregation strategy, the model can converge rapidly to the optimal performance.

arXiv comment16 pages, 3 figures, 4 tables

References in corpus (10)

• Sequence to Sequence Learning with Neural Networks
• Semi-Supervised Classification with Graph Convolutional Networks
• Inductive Representation Learning on Large Graphs
• Convolutional Sequence to Sequence Learning
• Graph Embedding Techniques, Applications, and Performance: A Survey
• Learning Convolutional Neural Networks for Graphs
• Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning
• FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling
• Learning Deep Generative Models of Graphs
• Learning to Execute

Cited by in corpus (43)

• Iterative Deep Graph Learning for Graph Neural Networks: Better and Robust Node Embeddings
• Graph Neural Networks: Taxonomy, Advances and Trends
• Heterogeneous Global Graph Neural Networks for Personalized Session-based Recommendation
• Improved Automatic Summarization of Subroutines via Attention to File Context
• On the Generalizability of Neural Program Models with respect to Semantic-Preserving Program Transformations
• Deep Learning-based Software Engineering: Progress, Challenges, and Opportunities
• Documentation Matters: Human-Centered AI System to Assist Data Science Code Documentation in Computational Notebooks
• Toward Subgraph-Guided Knowledge Graph Question Generation with Graph Neural Networks
• Graph Neural Networks for Natural Language Processing: A Survey
• Retrieval-Augmented Generation for Code Summarization via Hybrid GNN
• Reinforcement Learning Based Graph-to-Sequence Model for Natural Question Generation
• Deep Iterative and Adaptive Learning for Graph Neural Networks
• Graph Transformer for Graph-to-Sequence Learning
• Medical Entity Disambiguation Using Graph Neural Networks
• A Graph-to-Sequence Model for AMR-to-Text Generation
• GraphFlow: Exploiting Conversation Flow with Graph Neural Networks for Conversational Machine Comprehension
• Exploiting Rich Syntactic Information for Semantic Parsing with Graph-to-Sequence Model
• Cross-lingual Knowledge Graph Alignment via Graph Matching Neural Network
• Improving Question Answering over Incomplete KBs with Knowledge-Aware Reader
• One-Shot Relational Learning for Knowledge Graphs
• Graph-to-Tree Neural Networks for Learning Structured Input-Output Translation with Applications to Semantic Parsing and Math Word Problem
• Natural Question Generation with Reinforcement Learning Based Graph-to-Sequence Model
• Bidirectional Attentive Memory Networks for Question Answering over Knowledge Bases
• Graph-to-SFILES: Control structure prediction from process topologies using generative artificial intelligence
• Structural Neural Encoders for AMR-to-text Generation
• Coherent Comment Generation for Chinese Articles with a Graph-to-Sequence Model
• Reinforcement Learning Based Text Style Transfer without Parallel Training Corpus
• Graph-based Neural Sentence Ordering
• SQL-to-Text Generation with Graph-to-Sequence Model
• GraphSearchNet: Enhancing GNNs via Capturing Global Dependencies for Semantic Code Search
• HAConvGNN: Hierarchical Attention Based Convolutional Graph Neural Network for Code Documentation Generation in Jupyter Notebooks
• Permutation invariant graph-to-sequence model for template-free retrosynthesis and reaction prediction
• Ask Questions with Double Hints: Visual Question Generation with Answer-awareness and Region-reference
• Towards Graph Representation Learning in Emergent Communication
• Fast Incremental von Neumann Graph Entropy Computation: Theory, Algorithm, and Applications
• Graph Convolution for Multimodal Information Extraction from Visually Rich Documents
• Stronger Transformers for Neural Multi-Hop Question Generation
• DynGraph2Seq: Dynamic-Graph-to-Sequence Interpretable Learning for Health Stage Prediction in Online Health Forums
• A Multi-Modal Transformer-based Code Summarization Approach for Smart Contracts
• TAG : Type Auxiliary Guiding for Code Comment Generation
• Generating Diverse Descriptions from Semantic Graphs
• Towards Improving Interpretability of Language Model Generation through a Structured Knowledge Discovery Approach
• Understanding in Artificial Intelligence