Research map / Graph learning / Graph self-supervised learning
Return to Basics: Very Simple Graph Contrastive Learning via Noise Cancellation Principle
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a very simple graph contrastive learning method guided by a noise cancellation principle, aimed at heterophilic graphs, questioning whether complex augmentations, encoders, or negative sampling are needed.
Problem
Graph contrastive learning performs poorly on heterophilic graphs, where connected nodes often belong to different classes, and existing methods rely on complex augmentations, intricate encoders, or negative sampling.
Approach
A very simple GCL framework motivated by a noise cancellation principle, questioning whether the complexity of existing GCL methods is necessary for heterophilic graphs.
Experimental setup
Builds on: graph contrastive learning
Closest papers by meaning
- Does GCL Need a Large Number of Negative Samples? Enhancing Graph Contrastive Learning with Effective and Efficient Negative Sampling (AAAI 2025 · similarity 0.94)
- ReGCL: Rethinking Message Passing in Graph Contrastive Learning (AAAI 2024 · similarity 0.94)
- S3GCL: Spectral, Swift, Spatial Graph Contrastive Learning (ICML 2024 · similarity 0.93)
- Hierarchical Topology Isomorphism Expertise Embedded Graph Contrastive Learning (AAAI 2024 · similarity 0.93)
- Beyond Homophily: Graph Contrastive Learning with Macro-Micro Message Passing (AAAI 2025 · similarity 0.92)
- TopoGCL: Topological Graph Contrastive Learning (AAAI 2024 · similarity 0.92)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.