CoCo: A Coupled Contrastive Framework for Unsupervised Domain Adaptive Graph Classification

Nan Yin, Li Shen, Mengzhu Wang, Long Lan, Zeyu Ma, Chong Chen, Xian-Sheng Hua, Xiao Luo
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:40040-40053, 2023.

Abstract

Although graph neural networks (GNNs) have achieved impressive achievements in graph classification, they often need abundant task-specific labels, which could be extensively costly to acquire. A credible solution is to explore additional labeled graphs to enhance unsupervised learning on the target domain. However, how to apply GNNs to domain adaptation remains unsolved owing to the insufficient exploration of graph topology and the significant domain discrepancy. In this paper, we propose Coupled Contrastive Graph Representation Learning (CoCo), which extracts the topological information from coupled learning branches and reduces the domain discrepancy with coupled contrastive learning. CoCo contains a graph convolutional network branch and a hierarchical graph kernel network branch, which explore graph topology in implicit and explicit manners. Besides, we incorporate coupled branches into a holistic multi-view contrastive learning framework, which not only incorporates graph representations learned from complementary views for enhanced understanding, but also encourages the similarity between cross-domain example pairs with the same semantics for domain alignment. Extensive experiments on popular datasets show that our CoCo outperforms these competing baselines in different settings generally.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-yin23a, title = {{C}o{C}o: A Coupled Contrastive Framework for Unsupervised Domain Adaptive Graph Classification}, author = {Yin, Nan and Shen, Li and Wang, Mengzhu and Lan, Long and Ma, Zeyu and Chen, Chong and Hua, Xian-Sheng and Luo, Xiao}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {40040--40053}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/yin23a/yin23a.pdf}, url = {https://proceedings.mlr.press/v202/yin23a.html}, abstract = {Although graph neural networks (GNNs) have achieved impressive achievements in graph classification, they often need abundant task-specific labels, which could be extensively costly to acquire. A credible solution is to explore additional labeled graphs to enhance unsupervised learning on the target domain. However, how to apply GNNs to domain adaptation remains unsolved owing to the insufficient exploration of graph topology and the significant domain discrepancy. In this paper, we propose Coupled Contrastive Graph Representation Learning (CoCo), which extracts the topological information from coupled learning branches and reduces the domain discrepancy with coupled contrastive learning. CoCo contains a graph convolutional network branch and a hierarchical graph kernel network branch, which explore graph topology in implicit and explicit manners. Besides, we incorporate coupled branches into a holistic multi-view contrastive learning framework, which not only incorporates graph representations learned from complementary views for enhanced understanding, but also encourages the similarity between cross-domain example pairs with the same semantics for domain alignment. Extensive experiments on popular datasets show that our CoCo outperforms these competing baselines in different settings generally.} }
Endnote
%0 Conference Paper %T CoCo: A Coupled Contrastive Framework for Unsupervised Domain Adaptive Graph Classification %A Nan Yin %A Li Shen %A Mengzhu Wang %A Long Lan %A Zeyu Ma %A Chong Chen %A Xian-Sheng Hua %A Xiao Luo %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-yin23a %I PMLR %P 40040--40053 %U https://proceedings.mlr.press/v202/yin23a.html %V 202 %X Although graph neural networks (GNNs) have achieved impressive achievements in graph classification, they often need abundant task-specific labels, which could be extensively costly to acquire. A credible solution is to explore additional labeled graphs to enhance unsupervised learning on the target domain. However, how to apply GNNs to domain adaptation remains unsolved owing to the insufficient exploration of graph topology and the significant domain discrepancy. In this paper, we propose Coupled Contrastive Graph Representation Learning (CoCo), which extracts the topological information from coupled learning branches and reduces the domain discrepancy with coupled contrastive learning. CoCo contains a graph convolutional network branch and a hierarchical graph kernel network branch, which explore graph topology in implicit and explicit manners. Besides, we incorporate coupled branches into a holistic multi-view contrastive learning framework, which not only incorporates graph representations learned from complementary views for enhanced understanding, but also encourages the similarity between cross-domain example pairs with the same semantics for domain alignment. Extensive experiments on popular datasets show that our CoCo outperforms these competing baselines in different settings generally.
APA
Yin, N., Shen, L., Wang, M., Lan, L., Ma, Z., Chen, C., Hua, X. & Luo, X.. (2023). CoCo: A Coupled Contrastive Framework for Unsupervised Domain Adaptive Graph Classification. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:40040-40053 Available from https://proceedings.mlr.press/v202/yin23a.html.

Related Material