Self-Reinforced Graph Contrastive Learning

2025-05-19

AI Summary

In graph contrastive learning (GCL), low-quality positive samples often induce semantic and structural distortion. To address this, we propose Self-Augmented GCL (SA-GCL), the first framework to introduce a self-feedback paradigm for positive sample selection: it dynamically couples encoder representation capability growth with real-time positive sample quality assessment. We design a geometry-aware selector grounded in the manifold assumption, enabling iterative self-reinforcement. SA-GCL further integrates multi-strategy graph augmentation, probabilistic dynamic sampling, and optimized self-supervised contrastive loss. Extensive experiments on multiple graph-level classification benchmarks demonstrate significant improvements over state-of-the-art methods. As a plug-and-play module, SA-GCL consistently enhances model performance across diverse architectures, validating its cross-domain generalizability and robustness to structural perturbations and distributional shifts.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Sampling/Simulation-based Search
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Graphs serve as versatile data structures in numerous real-world domains-including social networks, molecular biology, and knowledge graphs-by capturing intricate relational information among entities. Among graph-based learning techniques, Graph Contrastive Learning (GCL) has gained significant attention for its ability to derive robust, self-supervised graph representations through the contrasting of positive and negative sample pairs. However, a critical challenge lies in ensuring high-quality positive pairs so that the intrinsic semantic and structural properties of the original graph are preserved rather than distorted. To address this issue, we propose SRGCL (Self-Reinforced Graph Contrastive Learning), a novel framework that leverages the model's own encoder to dynamically evaluate and select high-quality positive pairs. We designed a unified positive pair generator employing multiple augmentation strategies, and a selector guided by the manifold hypothesis to maintain the underlying geometry of the latent space. By adopting a probabilistic mechanism for selecting positive pairs, SRGCL iteratively refines its assessment of pair quality as the encoder's representational power improves. Extensive experiments on diverse graph-level classification tasks demonstrate that SRGCL, as a plug-in module, consistently outperforms state-of-the-art GCL methods, underscoring its adaptability and efficacy across various domains.

Problem

• Ensuring high-quality positive pairs in graph contrastive learning
• Preserving intrinsic semantic and structural graph properties
• Dynamically evaluating and selecting positive pairs adaptively

Innovation

• Dynamic positive pair selection via encoder
• Unified generator with multiple augmentations
• Manifold-guided selector preserves latent geometry