重新思考对比学习图自监督学习中的谱增强
Rethinking Spectral Augmentation for Contrast-based Graph Self-Supervised Learning
AI总结：
针对对比学习图自监督学习中谱增强的悖论，通过实证研究发现节点级随机边删除、图级随机边添加的简单边扰动在性能相当或更优的同时计算效率更高，理论分析也支持这一结论，为该领域提供新见解。
摘要：
近年来对比学习图自监督学习的兴起，促使人们对谱线索展开深入探索。谱增强通过修改图的谱特性（如特征值或特征向量）来提升模型性能，这一方法被广泛认可。然而，基于谱域看似矛盾假设的方法却都能显著提升学习性能，形成了有趣的悖论。通过大量实证研究，我们发现简单的边扰动（节点级自监督学习用随机边删除，图级自监督学习用随机边添加）在计算效率显著更高的同时，性能始终相当或更优。这表明复杂谱增强的计算开销可能无法证明其实际收益的合理性。我们对浅层GNN的InfoNCE损失界的理论分析进一步支持了这一观察结果。这些见解推动了该领域的发展，可能优化图自监督学习的理解与实现。
英文摘要：
The recent surge in contrast-based graph self-supervised learning has prominently featured an intensified exploration of spectral cues. Spectral augmentation, which involves modifying a graph's spectral properties such as eigenvalues or eigenvectors, is widely believed to enhance model performance. However, an intriguing paradox emerges, as methods grounded in seemingly conflicting assumptions regarding the spectral domain demonstrate notable enhancements in learning performance. Through extensive empirical studies, we find that simple edge perturbations - random edge dropping for node-level and random edge adding for graph-level self-supervised learning - consistently yield comparable or superior performance while being significantly more computationally efficient. This suggests that the computational overhead of sophisticated spectral augmentations may not justify their practical benefits. Our theoretical analysis of the InfoNCE loss bounds for shallow GNNs further supports this observation. The proposed insights represent a significant leap forward in the field, potentially refining the understanding and implementation of graph self-supervised learning.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
