GUIDE: Guidance-based Incremental Learning with Diffusion Models
This addresses the problem of maintaining performance on old tasks while learning new ones in AI systems, representing an incremental improvement over prior generative replay techniques.
The paper tackles catastrophic forgetting in continual learning by introducing GUIDE, a method that uses classifier guidance in diffusion models to generate rehearsal samples targeting forgotten information, resulting in significant reduction of forgetting and outperforming existing generative replay methods.
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.