Bringing Your Own View: Graph Contrastive Learning without Prefabricated Data Augmentations.

Yuning You, Tianlong Chen, Zhangyang Wang, Yang Shen
Author Information
  1. Yuning You: Texas A&M University.
  2. Tianlong Chen: University of Texas at Austin.
  3. Zhangyang Wang: University of Texas at Austin.
  4. Yang Shen: Texas A&M University.

Abstract

Self-supervision is recently surging at its new frontier of graph learning. It facilitates graph representations beneficial to downstream tasks; but its success could hinge on domain knowledge for handcraft or the often expensive trials and errors. Even its state-of-the-art representative, graph contrastive learning (GraphCL), is not completely free of those needs as GraphCL uses a prefabricated prior reflected by the ad-hoc manual selection of graph data augmentations. Our work aims at advancing GraphCL by answering the following questions: Accordingly, we have extended the prefabricated discrete prior in the augmentation set, to a learnable continuous prior in the parameter space of graph generators, assuming that graph priors , similar to the concept of image manifolds, can be learned by data generation. Furthermore, to form contrastive views without collapsing to trivial solutions due to the prior learnability, we have leveraged both principles of information minimization (InfoMin) and information bottleneck (InfoBN) to regularize the learned priors. Eventually, contrastive learning, InfoMin, and InfoBN are incorporated organically into one framework of bi-level optimization. Our principled and automated approach has proven to be competitive against the state-of-the-art graph self-supervision methods, including GraphCL, on benchmarks of small graphs; and shown even better generalizability on large-scale graphs, without resorting to human expertise or downstream validation. Our code is publicly released at https://github.com/Shen-Lab/GraphCL_Automated.

Keywords

References

  1. IEEE Trans Neural Netw Learn Syst. 2024 Feb;35(2):2747-2758 [PMID: 35895656]
  2. Adv Neural Inf Process Syst. 2022 Dec;35:1909-1922 [PMID: 37192934]
  3. Chem Sci. 2017 Oct 31;9(2):513-530 [PMID: 29629118]
  4. Bioinformatics. 2022 Sep 16;38(Suppl_2):ii68-ii74 [PMID: 36124802]
  5. Proc Mach Learn Res. 2020 Jul;119:10871-10880 [PMID: 33283198]
  6. J Chem Inf Model. 2015 Nov 23;55(11):2324-37 [PMID: 26479676]
  7. IEEE Trans Pattern Anal Mach Intell. 2023 Feb;45(2):2412-2429 [PMID: 35476575]

Grants

  1. R35 GM124952/NIGMS NIH HHS

Word Cloud

Created with Highcharts 10.0.0graphlearningcontrastiveGraphCLpriorinformationwithoutdownstreamstate-of-the-artprefabricateddatapriorslearnedminimizationInfoMinbottleneckInfoBNgraphsGraphSelf-supervisionrecentlysurgingnewfrontierfacilitatesrepresentationsbeneficialtaskssuccesshingedomainknowledgehandcraftoftenexpensivetrialserrorsEvenrepresentativecompletelyfreeneedsusesreflectedad-hocmanualselectionaugmentationsworkaimsadvancingansweringfollowingquestions:AccordinglyextendeddiscreteaugmentationsetlearnablecontinuousparameterspacegeneratorsassumingsimilarconceptimagemanifoldscangenerationFurthermoreformviewscollapsingtrivialsolutionsduelearnabilityleveragedprinciplesregularizeEventuallyincorporatedorganicallyoneframeworkbi-leveloptimizationprincipledautomatedapproachprovencompetitiveself-supervisionmethodsincludingbenchmarkssmallshownevenbettergeneralizabilitylarge-scaleresortinghumanexpertisevalidationcodepubliclyreleasedhttps://githubcom/Shen-Lab/GraphCL_AutomatedBringingView:ContrastiveLearningPrefabricatedDataAugmentationsgenerativemodel

Similar Articles

Cited By (4)