S-Prompts Learning with Pre-trained Transformers: An Occam's Razor for Domain Incremental Learning
This addresses the problem of forgetting in continual learning for AI systems, offering a scalable solution with minimal parameter increase, though it is incremental as it builds on existing prompting techniques.
The paper tackles catastrophic forgetting in domain incremental learning by proposing S-Prompting, a paradigm that learns prompts independently across domains using pre-trained transformers, achieving an average 30% relative improvement over state-of-the-art exemplar-free methods and surpassing methods using exemplars by about 6%.
State-of-the-art deep neural networks are still struggling to address the catastrophic forgetting problem in continual learning. In this paper, we propose one simple paradigm (named as S-Prompting) and two concrete approaches to highly reduce the forgetting degree in one of the most typical continual learning scenarios, i.e., domain increment learning (DIL). The key idea of the paradigm is to learn prompts independently across domains with pre-trained transformers, avoiding the use of exemplars that commonly appear in conventional methods. This results in a win-win game where the prompting can achieve the best for each domain. The independent prompting across domains only requests one single cross-entropy loss for training and one simple K-NN operation as a domain identifier for inference. The learning paradigm derives an image prompt learning approach and a novel language-image prompt learning approach. Owning an excellent scalability (0.03% parameter increase per domain), the best of our approaches achieves a remarkable relative improvement (an average of about 30%) over the best of the state-of-the-art exemplar-free methods for three standard DIL tasks, and even surpasses the best of them relatively by about 6% in average when they use exemplars. Source code is available at \url{https://github.com/iamwangyabin/S-Prompts}.