Why Does Dropping Edges Usually Outperform Adding Edges in Graph Contrastive Learning?
This work addresses a fundamental issue in graph representation learning for researchers and practitioners, offering theoretical insights and a practical method to enhance GCL performance, though it is incremental as it builds on existing GCL frameworks.
The paper tackles the problem of why dropping edges typically outperforms adding edges in graph contrastive learning (GCL) by introducing a new metric called Error Passing Rate (EPR) to quantify graph fit, and proposes EPAGCL, a novel GCL algorithm that uses both augmentations based on EPR weights, achieving improved performance validated on various real-world datasets.
Graph contrastive learning (GCL) has been widely used as an effective self-supervised learning method for graph representation learning. However, how to apply adequate and stable graph augmentation to generating proper views for contrastive learning remains an essential problem. Dropping edges is a primary augmentation in GCL while adding edges is not a common method due to its unstable performance. To our best knowledge, there is no theoretical analysis to study why dropping edges usually outperforms adding edges. To answer this question, we introduce a new metric, namely Error Passing Rate (EPR), to quantify how a graph fits the network. Inspired by the theoretical conclusions and the idea of positive-incentive noise, we propose a novel GCL algorithm, Error-PAssing-based Graph Contrastive Learning (EPAGCL), which uses both edge adding and edge dropping as its augmentations. To be specific, we generate views by adding and dropping edges based on the weights derived from EPR. Extensive experiments on various real-world datasets are conducted to validate the correctness of our theoretical analysis and the effectiveness of our proposed algorithm. Our code is available at: https://github.com/hyzhang98/EPAGCL.