Self-supervised Auxiliary Learning with Meta-paths for Heterogeneous Graphs

Abstract

Graph neural networks have shown superior performance in a wide range of applications providing a powerful representation of graph-structured data. Recent works show that the representation can be further improved by auxiliary tasks. However, the auxiliary tasks for heterogeneous graphs, which contain rich semantic information with various types of nodes and edges, have less explored in the literature. In this paper, to learn graph neural networks on heterogeneous graphs we propose a novel self-supervised auxiliary learning method using meta-paths, which are composite relations of multiple edge types. Our proposed method is learning to learn a primary task by predicting meta-paths as auxiliary tasks. This can be viewed as a type of meta-learning. The proposed method can identify an effective combination of auxiliary tasks and automatically balance them to improve the primary task. Our methods can be applied to any graph neural networks in a plug-in manner without manual labeling or additional data. The experiments demonstrate that the proposed method consistently improves the performance of link prediction and node classification on heterogeneous graphs.

arXiv commentNeural Information Processing Systems (NeurIPS), 2020

References in corpus (24)

• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Prototypical Networks for Few-shot Learning
• Inductive Representation Learning on Large Graphs
• DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
• Conditional Image Synthesis With Auxiliary Classifier GANs
• Simplifying Graph Convolutional Networks
• Graph Convolutional Matrix Completion
• Meta-SGD: Learning to Learn Quickly for Few-Shot Learning
• Interaction Networks for Learning about Objects, Relations and Physics
• Graph Transformer Networks
• FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling
• Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books
• Link Prediction Based on Graph Neural Networks
• Graph networks as learnable physics engines for inference and control
• Strategies for Pre-training Graph Neural Networks
• Fast Context Adaptation via Meta-Learning
• Graph-Bert: Only Attention is Needed for Learning Graph Representations
• NSML: Meet the MLaaS platform with a real-world case study
• Scene Graph Generation by Iterative Message Passing
• Self-Supervised Generalisation with Meta Auxiliary Learning
• Contrastive Learning for Image Captioning
• NSML: A Machine Learning Platform That Enables You to Focus on Your Models
• Learning What and Where to Transfer
• Pre-training Graph Neural Networks with Kernels

Cited by in corpus (6)

• Graph Self-Supervised Learning: A Survey
• Self-Supervised Hypergraph Transformer for Recommender Systems
• Machine Learning in High Energy Physics: A review of heavy-flavor jet tagging at the LHC
• Taxonomy Completion via Triplet Matching Network
• Hyperbolic Hypergraphs for Sequential Recommendation
• Pre-Trained Models for Heterogeneous Information Networks