Unifying Invariance and Spuriousity for Graph Out-of-Distribution via Probability of Necessity and Sufficiency

arXiv:2402.09165v1 Announce Type: new
Abstract: Graph Out-of-Distribution (OOD), requiring that models trained on biased data generalize to the unseen test data, has a massive of real-world applications. One of the most mainstream methods is to extract the invariant subgraph by aligning the original and augmented data with the help of environment augmentation. However, these solutions might lead to the loss or redundancy of semantic subgraph and further result in suboptimal generalization. To address this challenge, we propose a unified framework …

abstract applications arxiv augmented data biased data cs.lg data distribution extract graph massive probability test type via world