REP: Resource-Efficient Prompting for Rehearsal-Free Continual Learning

TL;DR

REP introduces a resource-efficient prompting method for rehearsal-free continual learning, significantly reducing computational and memory costs while maintaining high accuracy across multiple image classification tasks.

Contribution

It proposes novel techniques like prompt selection, adaptive token merging, and layer dropping to enhance efficiency in rehearsal-free continual learning.

Findings

Outperforms state-of-the-art methods in resource efficiency

Maintains high accuracy with reduced computational costs

Effective across multiple image classification datasets

Abstract

Recent rehearsal-free continual learning (CL) methods guided by prompts achieve strong performance on vision tasks with non-stationary data but remain resource-intensive, hindering real-world edge deployment. We introduce resource-efficient prompting (REP), which improves the computational and memory efficiency of prompt-based rehearsal-free continual learning methods while minimizing accuracy trade-offs. Our approach employs swift prompt selection to refine input data using a carefully provisioned model and introduces adaptive token merging (AToM) and adaptive layer dropping (ALD) for efficient prompt updates. AToM and ALD selectively skip data and model layers while preserving task-specific features during the learning of new tasks. Extensive experiments on multiple image classification datasets demonstrate REP's superior resource efficiency over state-of-the-art rehearsal-free CL methods.