Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

ARIEL: Adversarial Graph Contrastive Learning

About

Contrastive learning is an effective unsupervised method in graph representation learning, and the key component of contrastive learning lies in the construction of positive and negative samples. Previous methods usually utilize the proximity of nodes in the graph as the principle. Recently, the data-augmentation-based contrastive learning method has advanced to show great power in the visual domain, and some works extended this method from images to graphs. However, unlike the data augmentation on images, the data augmentation on graphs is far less intuitive and much harder to provide high-quality contrastive samples, which leaves much space for improvement. In this work, by introducing an adversarial graph view for data augmentation, we propose a simple but effective method, Adversarial Graph Contrastive Learning (ARIEL), to extract informative contrastive samples within reasonable constraints. We develop a new technique called information regularization for stable training and use subgraph sampling for scalability. We generalize our method from node-level contrastive learning to the graph level by treating each graph instance as a super-node. ARIEL consistently outperforms the current graph contrastive learning methods for both node-level and graph-level classification tasks on real-world datasets. We further demonstrate that ARIEL is more robust in the face of adversarial attacks.

Shengyu Feng, Baoyu Jing, Yada Zhu, Hanghang Tong• 2022

Related benchmarks

TaskDatasetResultRank
Node ClassificationCora (test)
Mean Accuracy87.3
861
Node ClassificationCiteseer (test)
Accuracy0.7953
824
Node ClassificationPubMed (test)
Accuracy86.42
546
Node ClassificationChameleon (test)
Mean Accuracy64.53
297
Node ClassificationCornell (test)
Mean Accuracy70.7
274
Node ClassificationTexas (test)
Mean Accuracy76.19
269
Node ClassificationSquirrel (test)
Mean Accuracy42.42
267
Node ClassificationWisconsin (test)
Mean Accuracy71.15
239
Node ClassificationActor (test)
Mean Accuracy0.3768
237
Node ClassificationCora Poisoned
Accuracy84.8
17
Showing 10 of 17 rows

Other info

Follow for update