Read Between the Layers: Leveraging Intra-Layer Representations for Rehearsal-Free Continual Learning with Pre-Trained Models

arXiv:2312.08888v2 Announce Type: replace
Abstract: We address the Continual Learning (CL) problem, wherein a model must learn a sequence of tasks from non-stationary distributions while preserving prior knowledge upon encountering new experiences. With the advancement of foundation models, CL research has pivoted from the initial learning-from-scratch paradigm towards utilizing generic features from large-scale pre-training. However, existing approaches to CL with pre-trained models primarily focus on separating class-specific features from the final representation layer and neglect the potential of intermediate representations …

abstract advancement arxiv continual cs.cv cs.lg foundation free from-scratch knowledge layer learn pre-trained models prior research scratch tasks type