D-CBRS: Accounting For Intra-Class Diversity in Continual Learning

TL;DR

D-CBRS introduces a novel memory sampling method for continual learning that considers intra-class diversity, significantly improving performance on imbalanced and diverse datasets.

Contribution

The paper proposes D-CBRS, a new algorithm that accounts for intra-class diversity in memory sampling, enhancing continual learning with imbalanced data.

Findings

D-CBRS outperforms existing methods on datasets with high intra-class diversity.

Incorporating intra-class diversity improves memory efficiency and learning stability.

The approach reduces forgetting in class-imbalanced continual learning scenarios.

Abstract

Continual learning -- accumulating knowledge from a sequence of learning experiences -- is an important yet challenging problem. In this paradigm, the model's performance for previously encountered instances may substantially drop as additional data are seen. When dealing with class-imbalanced data, forgetting is further exacerbated. Prior work has proposed replay-based approaches which aim at reducing forgetting by intelligently storing instances for future replay. Although Class-Balancing Reservoir Sampling (CBRS) has been successful in dealing with imbalanced data, the intra-class diversity has not been accounted for, implicitly assuming that each instance of a class is equally informative. We present Diverse-CBRS (D-CBRS), an algorithm that allows us to consider within class diversity when storing instances in the memory. Our results show that D-CBRS outperforms state-of-the-art memory management continual learning algorithms on data sets with considerable intra-class diversity.