Memory-efficient Continual Learning with Prototypical Exemplar Condensation

TL;DR

This paper introduces a memory-efficient continual learning method that synthesizes prototypical exemplars and uses augmentation to reduce memory needs while maintaining high performance.

Contribution

It proposes a novel exemplar condensation technique and augmentation mechanism to improve memory efficiency and performance in continual learning.

Findings

Achieves superior performance on benchmark datasets.

Reduces memory footprint compared to existing methods.

Effective in large-scale, multi-task scenarios.

Abstract

Rehearsal-based continual learning (CL) mitigates catastrophic forgetting by maintaining a subset of samples from previous tasks for replay. Existing studies primarily focus on optimizing memory storage through coreset selection strategies. While these methods are effective, they typically require storing a substantial number of samples per class (SPC), often exceeding 20, to maintain satisfactory performance. In this work, we propose to further compress the memory footprint by synthesizing and storing prototypical exemplars, which can form representative prototypes when passed through the feature extractor. Owing to their representative nature, these exemplars enable the model to retain previous knowledge using only a small number of samples while preserving privacy. Moreover, we introduce a perturbation-based augmentation mechanism that generates synthetic variants of previous data during training, thereby enhancing CL performance. Extensive evaluations on widely used benchmark datasets and settings demonstrate that the proposed algorithm achieves superior performance compared to existing baselines, particularly in scenarios involving large-scale datasets and a high number of tasks.