.png){kind=icon}
Return
Information bottleneck based graph structural learning for OOD generalization
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ipm.2026.104691.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• We propose a graph structural learning method for graph OOD generalization. • We design three modules for mitigating spurious correlations and domain shifts. • We conduct extensive experiments to demonstrate our effectiveness.
Keywords:
graph structural learning
out-of-distribution generalization
information bottleneck
spurious correlations
domain shift
Journal
I
IF:
6.9
Papers:
533
Citations:
0
Organization
Cited Papers
ShareSaveShareSaveShareSaveCOOL: A Conjoint Perspective on Spatio-Temporal Graph Neural Network for Traffic Forecasting.png)
INFORMATION FUSION
OAAI
INFORMATION FUSIONIF15.5
OAAIShareSaveSpatio-temporal fusion graph convolutional network for traffic flow forecastingINFORMATION FUSION
INFORMATION FUSIONIF15.5
PREAI
ShareSave
View more