Improving Forward Compatibility in Class Incremental Learning by Increasing Representation Rank and Feature Richness

arXiv:2403.15517v1 Announce Type: new
Abstract: Class Incremental Learning (CIL) constitutes a pivotal subfield within continual learning, aimed at enabling models to progressively learn new classification tasks while retaining knowledge obtained from prior tasks. Although previous studies have predominantly focused on backward compatible approaches to mitigate catastrophic forgetting, recent investigations have introduced forward compatible methods to enhance performance on novel tasks and complement existing backward compatible methods. In this study, we introduce an effective-Rank based Feature Richness enhancement (RFR) method, designed …

abstract arxiv class classification continual cs.cv cs.lg enabling feature improving incremental knowledge learn pivotal prior representation studies tasks type