notesum.ai
Published at October 23Mitigating Graph Covariate Shift via Score-based Out-of-distribution Augmentation
cs.LG
cs.AI
Released Date: October 23, 2024
Authors: Bohan Wang1, Yurui Chang1, Lu Lin1
Aff.: 1The Pennsylvania State University

| Type | Method | Motif | CMNIST | Molhiv | GOOD-SST2 | |||
|---|---|---|---|---|---|---|---|---|
| base | size | color | scaffold | size | length | |||
| General Generalization | ERM | |||||||
| IRM | ||||||||
| GroupDRO | ||||||||
| VREx | ||||||||
| DANN | ||||||||
| Deep Coral | ||||||||
| Graph Generalization | DIR | |||||||
| GSAT | ||||||||
| CIGA | ||||||||
| Graph Augmentation | DropNode | |||||||
| DropEdge | ||||||||
| MaskFeature | ||||||||
| FLAG | ||||||||
| M-Mixup | ||||||||
| G-Mixup | ||||||||
| AIA | ||||||||
| OODA(Ours) | ||||||||