TY - GEN
T1 - Supervised Graph Contrastive Learning for Few-Shot Node Classification
AU - Tan, Zhen
AU - Ding, Kaize
AU - Guo, Ruocheng
AU - Liu, Huan
N1 - Funding Information: Acknowledgments. This work is partially supported by Army Research Office (ARO) W911NF2110030 and Army Research Lab (ARL) W911NF2020124. Publisher Copyright: © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2023
Y1 - 2023
N2 - Graphs present in many real-world applications, such as financial fraud detection, commercial recommendation, and social network analysis. But given the high cost of graph annotation or labeling, we face a severe graph label-scarcity problem, i.e., a graph might have a few labeled nodes. One example of such a problem is the so-called few-shot node classification. A predominant approach to this problem resorts to episodic meta-learning. In this work, we challenge the status quo by asking a fundamental question whether meta-learning is a must for few-shot node classification tasks. We propose a new and simple framework under the standard few-shot node classification setting as an alternative to meta-learning to learn an effective graph encoder. The framework consists of supervised graph contrastive learning with novel mechanisms for data augmentation, subgraph encoding, and multi-scale contrast on graphs. Extensive experiments on three benchmark datasets (CoraFull, Reddit, Ogbn) show that the new framework significantly outperforms state-of-the-art meta-learning based methods.
AB - Graphs present in many real-world applications, such as financial fraud detection, commercial recommendation, and social network analysis. But given the high cost of graph annotation or labeling, we face a severe graph label-scarcity problem, i.e., a graph might have a few labeled nodes. One example of such a problem is the so-called few-shot node classification. A predominant approach to this problem resorts to episodic meta-learning. In this work, we challenge the status quo by asking a fundamental question whether meta-learning is a must for few-shot node classification tasks. We propose a new and simple framework under the standard few-shot node classification setting as an alternative to meta-learning to learn an effective graph encoder. The framework consists of supervised graph contrastive learning with novel mechanisms for data augmentation, subgraph encoding, and multi-scale contrast on graphs. Extensive experiments on three benchmark datasets (CoraFull, Reddit, Ogbn) show that the new framework significantly outperforms state-of-the-art meta-learning based methods.
KW - Few-shot learning
KW - Graph Neural Networks
KW - Graph contrastive learning
UR - https://www.scopus.com/pages/publications/85151059097
UR - https://www.scopus.com/pages/publications/85151059097#tab=citedBy
U2 - 10.1007/978-3-031-26390-3_24
DO - 10.1007/978-3-031-26390-3_24
M3 - Conference contribution
SN - 9783031263897
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 394
EP - 411
BT - Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2022, Proceedings
A2 - Amini, Massih-Reza
A2 - Canu, Stéphane
A2 - Fischer, Asja
A2 - Guns, Tias
A2 - Kralj Novak, Petra
A2 - Tsoumakas, Grigorios
PB - Springer Science and Business Media Deutschland GmbH
T2 - 22nd Joint European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2022
Y2 - 19 September 2022 through 23 September 2022
ER -