Generating Counterfactual Hard Negative Samples for Graph Contrastive Learning

Yang, H; Chen, H; Zhang, S; Sun, X; Li, Q; Zhao, X; Xu, G

Generating Counterfactual Hard Negative Samples for Graph Contrastive Learning

Yang, H Chen, H Zhang, S Sun, X Li, Q Zhao, X Xu, G

Permalink

Publisher:: Association for Computing Machinery (ACM)
Publication Type:: Conference Proceeding
Citation:: ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023, 2023, pp. 621-629
Issue Date:: 2023-04-30

Recently Added

	Filename	Description	Size
	Generating Counterfactual Hard Negative Samples.pdf	Submitted version	1.03 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is new to OPUS and is not currently available.

Full metadata record

Field	Value	Language
dc.contributor.author	Yang, H
dc.contributor.author	Chen, H
dc.contributor.author	Zhang, S
dc.contributor.author	Sun, X
dc.contributor.author	Li, Q
dc.contributor.author	Zhao, X
dc.contributor.author	Xu, G https://orcid.org/0000-0003-4493-6663
dc.date.accessioned	2024-03-28T03:42:51Z
dc.date.available	2024-03-28T03:42:51Z
dc.date.issued	2023-04-30
dc.identifier.citation	ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023, 2023, pp. 621-629
dc.identifier.isbn	9781450394161
dc.identifier.uri	http://hdl.handle.net/10453/177336
dc.description.abstract	Graph contrastive learning has emerged as a powerful unsupervised graph representation learning tool. The key to the success of graph contrastive learning is to acquire high-quality positive and negative samples as contrasting pairs to learn the underlying structural semantics of the input graph. Recent works usually sample negative samples from the same training batch with the positive samples or from an external irrelevant graph. However, a significant limitation lies in such strategies: the unavoidable problem of sampling false negative samples. In this paper, we propose a novel method to utilize Counterfactual mechanism to generate artificial hard negative samples for Graph Contrastive learning, namely CGC. We utilize a counterfactual mechanism to produce hard negative samples, ensuring that the generated samples are similar but have labels that differ from the positive sample. The proposed method achieves satisfying results on several datasets. It outperforms some traditional unsupervised graph learning methods and some SOTA graph contrastive learning methods. We also conducted some supplementary experiments to illustrate the proposed method, including the performances of CGC with different hard negative samples and evaluations for hard negative samples generated with different similarity measurements. The implementation code is available online to ease reproducibility1.
dc.language	en
dc.publisher	Association for Computing Machinery (ACM)
dc.relation	http://purl.org/au-research/grants/arc/LP170100891
dc.relation	http://purl.org/au-research/grants/arc/DP200101374
dc.relation	http://purl.org/au-research/grants/arc/DP220103717
dc.relation	http://purl.org/au-research/grants/arc/LE220100078
dc.relation.ispartof	ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023
dc.relation.ispartof	Proceedings of the ACM Web Conference 2023
dc.relation.isbasedon	10.1145/3543507.3583499
dc.rights	info:eu-repo/semantics/restrictedAccess
dc.title	Generating Counterfactual Hard Negative Samples for Graph Contrastive Learning
dc.type	Conference Proceeding
pubs.organisational-group	University of Technology Sydney
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	University of Technology Sydney/Strength - AAI - Advanced Analytics Institute Research Centre
pubs.organisational-group	University of Technology Sydney/Faculty of Engineering and Information Technology/School of Computer Science
utslib.copyright.status	recently_added	*
dc.date.updated	2024-03-28T03:42:50Z
pubs.publication-status	Published

Abstract:

Graph contrastive learning has emerged as a powerful unsupervised graph representation learning tool. The key to the success of graph contrastive learning is to acquire high-quality positive and negative samples as contrasting pairs to learn the underlying structural semantics of the input graph. Recent works usually sample negative samples from the same training batch with the positive samples or from an external irrelevant graph. However, a significant limitation lies in such strategies: the unavoidable problem of sampling false negative samples. In this paper, we propose a novel method to utilize Counterfactual mechanism to generate artificial hard negative samples for Graph Contrastive learning, namely CGC. We utilize a counterfactual mechanism to produce hard negative samples, ensuring that the generated samples are similar but have labels that differ from the positive sample. The proposed method achieves satisfying results on several datasets. It outperforms some traditional unsupervised graph learning methods and some SOTA graph contrastive learning methods. We also conducted some supplementary experiments to illustrate the proposed method, including the performances of CGC with different hard negative samples and evaluations for hard negative samples generated with different similarity measurements. The implementation code is available online to ease reproducibility1.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/177336