Timezone: »

From Canonical Correlation Analysis to Self-supervised Graph Neural Networks
Hengrui Zhang · Qitian Wu · Junchi Yan · David P Wipf · Philip S Yu

Thu Dec 09 12:30 AM -- 02:00 AM (PST) @ Virtual #None

We introduce a conceptually simple yet effective model for self-supervised representation learning with graph data. It follows the previous methods that generate two views of an input graph through data augmentation. However, unlike contrastive methods that focus on instance-level discrimination, we optimize an innovative feature-level objective inspired by classical Canonical Correlation Analysis. Compared with other works, our approach requires none of the parameterized mutual information estimator, additional projector, asymmetric structures, and most importantly, negative samples which can be costly. We show that the new objective essentially 1) aims at discarding augmentation-variant information by learning invariant representations, and 2) can prevent degenerated solutions by decorrelating features in different dimensions. Our theoretical analysis further provides an understanding for the new objective which can be equivalently seen as an instantiation of the Information Bottleneck Principle under the self-supervised setting. Despite its simplicity, our method performs competitively on seven public graph datasets.

Author Information

Hengrui Zhang (University of Illinois at Chicago)
Qitian Wu (Shanghai Jiao Tong University)
Junchi Yan (Shanghai Jiao Tong University)
David P Wipf (AWS)
Philip S Yu (UIC)

More from the Same Authors