Knowledge Graph Contrastive Learning Based on Relation-Symmetrical Structure

Abstract

Knowledge graph embedding (KGE) aims at learning powerful representations to benefit various artificial intelligence applications. Meanwhile, contrastive learning has been widely leveraged in graph learning as an effective mechanism to enhance the discriminative capacity of the learned representations. However, the complex structures of KG make it hard to construct appropriate contrastive pairs. Only a few attempts have integrated contrastive learning strategies with KGE. But, most of them rely on language models ( e.g., Bert) for contrastive pair construction instead of fully mining information underlying the graph structure, hindering expressive ability. Surprisingly, we find that the entities within a relational symmetrical structure are usually similar and correlated. To this end, we propose a knowledge graph contrastive learning framework based on relation-symmetrical structure, KGE-SymCL, which mines symmetrical structure information in KGs to enhance the discriminative ability of KGE models. Concretely, a plug-and-play approach is proposed by taking entities in the relation-symmetrical positions as positive pairs. Besides, a self-supervised alignment loss is designed to pull together positive pairs. Experimental results on link prediction and entity classification datasets demonstrate that our KGE-SymCL can be easily adopted to various KGE models for performance improvements. Moreover, extensive experiments show that our model could outperform other state-of-the-art baselines.

This work has been accepted by IEEE for publication. Early access in IEEE Transactions on Knowledge and Data Engineering

References in corpus (21)

• A Simple Framework for Contrastive Learning of Visual Representations
• Bootstrap your own latent: A new approach to self-supervised Learning
• A Survey on Knowledge Graphs: Representation, Acquisition and Applications
• Graph Contrastive Learning with Adaptive Augmentation
• Graph Contrastive Learning with Augmentations
• Barlow Twins: Self-Supervised Learning via Redundancy Reduction
• Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learning
• Knowledge Graph Contrastive Learning for Recommendation
• Deep Graph Contrastive Representation Learning
• Contrastive Multi-View Representation Learning on Graphs
• SimGRACE: A Simple Framework for Graph Contrastive Learning without Data Augmentation
• Hard Negative Mixing for Contrastive Learning
• Adversarial Graph Augmentation to Improve Graph Contrastive Learning
• Graph Contrastive Learning Automated
• Graph-MLP: Node Classification without Message Passing in Graph
• ProGCL: Rethinking Hard Negative Mining in Graph Contrastive Learning
• Duality-Induced Regularizer for Tensor Factorization Based Knowledge Graph Completion
• HousE: Knowledge Graph Embedding with Householder Parameterization
• Learning Deep Graph Representations via Convolutional Neural Networks
• Improved Dual Correlation Reduction Network
• R-GCN: The R Could Stand for Random

Cited by in corpus (5)

• Self-Supervised Temporal Graph learning with Temporal and Structural Intensity Alignment
• Parallel Knowledge Enhancement based Framework for Multi-behavior Recommendation
• Learning to Denoise Biomedical Knowledge Graph for Robust Molecular Interaction Prediction
• MORE: A Multimodal Object-Entity Relation Extraction Dataset with a Benchmark Evaluation
• Towards Mitigating Dimensional Collapse of Representations in Collaborative Filtering