GUIDE: Guidance-based Incremental Learning with Diffusion Models

TL;DR

GUIDE is a continual learning method that uses classifier-guided diffusion models to generate targeted rehearsal samples, effectively reducing catastrophic forgetting and outperforming existing generative replay techniques.

Contribution

It introduces a guidance-based sampling strategy in diffusion models for rehearsal, bridging the gap between generative and buffer-based continual learning methods.

Findings

GUIDE significantly reduces catastrophic forgetting.

Outperforms state-of-the-art continual learning methods.

Generates more relevant rehearsal samples for past tasks.

Abstract

We introduce GUIDE, a novel continual learning approach that directs diffusion models to rehearse samples at risk of being forgotten. Existing generative strategies combat catastrophic forgetting by randomly sampling rehearsal examples from a generative model. Such an approach contradicts buffer-based approaches where sampling strategy plays an important role. We propose to bridge this gap by incorporating classifier guidance into the diffusion process to produce rehearsal examples specifically targeting information forgotten by a continuously trained model. This approach enables the generation of samples from preceding task distributions, which are more likely to be misclassified in the context of recently encountered classes. Our experimental results show that GUIDE significantly reduces catastrophic forgetting, outperforming conventional random sampling approaches and surpassing recent state-of-the-art methods in continual learning with generative replay.