TY - GEN
T1 - SGOOD
T2 - 33rd ACM International Conference on Information and Knowledge Management, CIKM 2024
AU - Ding, Zhihao
AU - Shi, Jieming
AU - Shen, Shiqi
AU - Shang, Xuequn
AU - Cao, Jiannong
AU - Wang, Zhipeng
AU - Gong, Zhi
N1 - Publisher Copyright:
© 2024 Owner/Author.
PY - 2024/10/21
Y1 - 2024/10/21
N2 - Graph-level representation learning is important in a wide range of applications. Existing graph-level models are generally built on i.i.d. assumption for both training and testing graphs. However, in an open world, models can encounter out-of-distribution (OOD) testing graphs that are from different distributions unknown during training. A trustworthy model should be able to detect OOD graphs to avoid unreliable predictions, while producing accurate in-distribution (ID) predictions. To achieve this, we present SGOOD, a novel graph-level OOD detection framework. We find that substructure differences commonly exist between ID and OOD graphs, and design SGOOD with a series of techniques to encode task-agnostic substructures for effective OOD detection. Specifically, we build a super graph of substructures for every graph, and develop a two-level graph encoding pipeline that works on both original graphs and super graphs to obtain substructure-enhanced graph representations. We then devise substructure-preserving graph augmentation techniques to further capture more substructure semantics of ID graphs. Extensive experiments against 11 competitors on numerous graph datasets demonstrate the superiority of SGOOD, often surpassing existing methods by a significant margin. The code is available at https://github.com/TommyDzh/SGOOD.
AB - Graph-level representation learning is important in a wide range of applications. Existing graph-level models are generally built on i.i.d. assumption for both training and testing graphs. However, in an open world, models can encounter out-of-distribution (OOD) testing graphs that are from different distributions unknown during training. A trustworthy model should be able to detect OOD graphs to avoid unreliable predictions, while producing accurate in-distribution (ID) predictions. To achieve this, we present SGOOD, a novel graph-level OOD detection framework. We find that substructure differences commonly exist between ID and OOD graphs, and design SGOOD with a series of techniques to encode task-agnostic substructures for effective OOD detection. Specifically, we build a super graph of substructures for every graph, and develop a two-level graph encoding pipeline that works on both original graphs and super graphs to obtain substructure-enhanced graph representations. We then devise substructure-preserving graph augmentation techniques to further capture more substructure semantics of ID graphs. Extensive experiments against 11 competitors on numerous graph datasets demonstrate the superiority of SGOOD, often surpassing existing methods by a significant margin. The code is available at https://github.com/TommyDzh/SGOOD.
KW - graph classification
KW - out-of-distribution detection
KW - reliability
KW - trustworthy model
UR - http://www.scopus.com/inward/record.url?scp=85210026583&partnerID=8YFLogxK
U2 - 10.1145/3627673.3679710
DO - 10.1145/3627673.3679710
M3 - 会议稿件
AN - SCOPUS:85210026583
T3 - International Conference on Information and Knowledge Management, Proceedings
SP - 467
EP - 476
BT - CIKM 2024 - Proceedings of the 33rd ACM International Conference on Information and Knowledge Management
PB - Association for Computing Machinery
Y2 - 21 October 2024 through 25 October 2024
ER -