DyTox: Transformers for Continual Learning with DYnamic TOken eXpansion
This addresses the problem of scaling continual learning to many tasks for AI systems, though it is incremental as it builds on dynamic architecture trends.
The paper tackles catastrophic forgetting in continual learning by proposing DyTox, a transformer architecture with dynamic token expansion that shares encoder/decoder across tasks, achieving state-of-the-art results on ImageNet100 and ImageNet1000 with fewer parameters than existing dynamic frameworks.
Deep network architectures struggle to continually learn new tasks without forgetting the previous tasks. A recent trend indicates that dynamic architectures based on an expansion of the parameters can reduce catastrophic forgetting efficiently in continual learning. However, existing approaches often require a task identifier at test-time, need complex tuning to balance the growing number of parameters, and barely share any information across tasks. As a result, they struggle to scale to a large number of tasks without significant overhead. In this paper, we propose a transformer architecture based on a dedicated encoder/decoder framework. Critically, the encoder and decoder are shared among all tasks. Through a dynamic expansion of special tokens, we specialize each forward of our decoder network on a task distribution. Our strategy scales to a large number of tasks while having negligible memory and time overheads due to strict control of the parameters expansion. Moreover, this efficient strategy doesn't need any hyperparameter tuning to control the network's expansion. Our model reaches excellent results on CIFAR100 and state-of-the-art performances on the large-scale ImageNet100 and ImageNet1000 while having less parameters than concurrent dynamic frameworks.