Enhancing Intrinsic Features for Debiasing via Investigating Class-Discerning Common Attributes in Bias-Contrastive Pair

arXiv:2404.19250v1 Announce Type: new
Abstract: In the image classification task, deep neural networks frequently rely on bias attributes that are spuriously correlated with a target class in the presence of dataset bias, resulting in degraded performance when applied to data without bias attributes. The task of debiasing aims to compel classifiers to learn intrinsic attributes that inherently define a target class rather than focusing on bias attributes. While recent approaches mainly focus on emphasizing the learning of data samples without …

abstract arxiv bias class classification cs.cv data dataset features image intrinsic networks neural networks performance type via