Graph contrastive learning with consistency regularization

Soohong Lee, Sangho Lee, Jaehwan Lee, Woojin Lee, Youngdoo Son

Abstract: Highlights•We propose a graph contrastive learning method to address the class collision problem.•We suggest a novel loss function that adds a consistency regularization term.•Our method obtains effective graph representation reflecting intra-class similarity.•Experimental results show that our method outperforms previous state-of-the-art methods.