Spuriousness-Aware Meta-Learning for Learning Robust Classifiers

arXiv:2406.10742v1 Announce Type: new
Abstract: Spurious correlations are brittle associations between certain attributes of inputs and target variables, such as the correlation between an image background and an object class. Deep image classifiers often leverage them for predictions, leading to poor generalization on the data where the correlations do not hold. Mitigating the impact of spurious correlations is crucial towards robust model generalization, but it often requires annotations of the spurious correlations in data -- a strong assumption in practice. …

abstract arxiv attributes class classifiers correlation correlations cs.cv data image impact inputs meta meta-learning object predictions robust them type variables