Active Learning for Continual Learning: Keeping the Past Alive in the Present

TL;DR

This paper introduces AccuACL, a novel active continual learning method that uses Fisher information to select informative samples, effectively balancing learning new tasks and retaining past knowledge, thereby reducing forgetting.

Contribution

It proposes a Fisher information-based sample selection strategy for ACL, addressing scalability and balancing learning and retention, which outperforms existing AL methods.

Findings

Significantly improves average accuracy by 23.8%.

Reduces forgetting by 17.0%.

Outperforms baseline methods across various CL algorithms.

Abstract

Continual learning (CL) enables deep neural networks to adapt to ever-changing data distributions. In practice, there may be scenarios where annotation is costly, leading to active continual learning (ACL), which performs active learning (AL) for the CL scenarios when reducing the labeling cost by selecting the most informative subset is preferable. However, conventional AL strategies are not suitable for ACL, as they focus solely on learning the new knowledge, leading to catastrophic forgetting of previously learned tasks. Therefore, ACL requires a new AL strategy that can balance the prevention of catastrophic forgetting and the ability to quickly learn new tasks. In this paper, we propose AccuACL, Accumulated informativeness-based Active Continual Learning, by the novel use of the Fisher information matrix as a criterion for sample selection, derived from a theoretical analysis of the Fisher-optimality preservation properties within the framework of ACL, while also addressing the scalability issue of Fisher information-based AL. Extensive experiments demonstrate that AccuACL significantly outperforms AL baselines across various CL algorithms, increasing the average accuracy and forgetting by 23.8% and 17.0%, respectively, on average.