{"type":"ScholarlyArticle","title":"OS-GCL: A One-Shot Learner in Graph Contrastive Learning","authors":[{"name":"Cheng Ji","slug":"cheng-ji","given":"Cheng","family":"Ji","affiliation":""},{"name":"Chenrui He","slug":"chenrui-he","given":"Chenrui","family":"He","affiliation":""},{"name":"Qian Li","slug":"qian-li","given":"Qian","family":"Li","affiliation":""},{"name":"Qingyun Sun","slug":"qingyun-sun","given":"Qingyun","family":"Sun","affiliation":""},{"name":"Xingcheng Fu","slug":"xingcheng-fu","given":"Xingcheng","family":"Fu","affiliation":""},{"name":"Jianxin Li","slug":"jianxin-li","given":"Jianxin","family":"Li","affiliation":""}],"abstract":"Graph contrastive learning (GCL) enhances the self-supervised learning capacity for graph representation learning. Nevertheless, the previous research has neglected to consider one fundamental nature of GCL -- graph contrastive learning operates as a one-shot learner, guided by the widely utilized noise contrastive estimation (e.g., the InfoNCE loss). Theoretically, to initially investigate the factors that contribute to the one-shot learner essence, we analyze the InfoNCE-based objective and derive its equivalent form of the softmax-based cross-entropy function. It is concluded that the InfoNCE-based GCL is determined to be a (2n-1)-way 1-shot classifier (n is the number of nodes). In this particular context, each sample is indicative of a unique ideational class, and each class has only one sample. Consequently, the one-shot learning nature of GCL leads to the issue of the limited self-supervised signal. To further address the above issue, we propose a One-Shot Learner in Graph Contrastive Learning (OS-GCL). Firstly, we estimate the potential probability distributions of the deterministic node features and discrete graph topology. Secondly, we develop a probabilistic message-passing mechanism to propagate probability (of feature) on probability (of topology). Thirdly, we propose the ProbNCE loss functions to contrast distributions. Extensive experimental results demonstrate the superiority of OS-GCL. To the best of our knowledge, this is the first study to examine the one-shot learning essence and the limited self-supervised signal issue of GCL.","doi":"10.24963/ijcai.2025/330","url":"https://ijcai.org/proceedings/2025/330","pdf":"https://ijcai.org/proceedings/2025/0330.pdf","markdown":"https://ijcai.org/proceedings/2025/330.md","bibtex":"https://ijcai.org/proceedings/2025/bibtex/330","volume":{"slug":"2025","title":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2025","url":"https://ijcai.org/proceedings/2025/","year":2025,"location":"Montreal, Canada","dates":"16-22 August 2025","editors":"James Kwok","isbn":"978-1-956792-06-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2025"},"section":"Data Mining","track":"Main Track","number":330,"pages":{"first":2964,"last":2972},"keywords":["Data Mining: DM: Mining graphs"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}