Toward Sustainable Continual Learning: Detection and Knowledge Repurposing of Similar Tasks
This addresses the scalability issue in continual learning for AI systems by reducing memory growth, though it is incremental as it builds on existing methods.
The paper tackles the problem of prohibitive memory expansion in continual learning by introducing a framework that detects and reuses knowledge from similar tasks, achieving competitive performance on CIFAR10, CIFAR100, and EMNIST benchmarks.
Most existing works on continual learning (CL) focus on overcoming the catastrophic forgetting (CF) problem, with dynamic models and replay methods performing exceptionally well. However, since current works tend to assume exclusivity or dissimilarity among learning tasks, these methods require constantly accumulating task-specific knowledge in memory for each task. This results in the eventual prohibitive expansion of the knowledge repository if we consider learning from a long sequence of tasks. In this work, we introduce a paradigm where the continual learner gets a sequence of mixed similar and dissimilar tasks. We propose a new continual learning framework that uses a task similarity detection function that does not require additional learning, with which we analyze whether there is a specific task in the past that is similar to the current task. We can then reuse previous task knowledge to slow down parameter expansion, ensuring that the CL system expands the knowledge repository sublinearly to the number of learned tasks. Our experiments show that the proposed framework performs competitively on widely used computer vision benchmarks such as CIFAR10, CIFAR100, and EMNIST.