Sequencing to Mitigate Catastrophic Forgetting in Continual Learning
Hesham G. Moussa, Aroosa Hameed, Arashmid Akhavain
TL;DR
The paper investigates how the order in which tasks are presented to a continual learning model affects catastrophic forgetting. It introduces NWOT, a zero-shot NAS-inspired method, to predict the most informative next dataset in a sequence, and extends it with Activation Interval Dropout (AID) to handle non-IID data. The approach is evaluated on DomainNet with IID and non-IID scenarios, demonstrating that intelligent sequencing can substantially reduce forgetting, especially when combined with standard CL techniques like EWC. The work highlights sequencing as a viable, lightweight lever to improve continual learning robustness and suggests avenues for further improvements, including more sophisticated prediction strategies and integration with broader AI systems.
Abstract
To cope with real-world dynamics, an intelligent system needs to incrementally acquire, update, and exploit knowledge throughout its lifetime. This ability, known as Continual learning, provides a foundation for AI systems to develop themselves adaptively. Catastrophic forgetting is a major challenge to the progress of Continual Learning approaches, where learning a new task usually results in a dramatic performance drop on previously learned ones. Many approaches have emerged to counteract the impact of CF. Most of the proposed approaches can be categorized into five classes: replay-based, regularization-based, optimization-based, representation-based, and architecture-based. In this work, we approach the problem from a different angle, specifically by considering the optimal sequencing of tasks as they are presented to the model. We investigate the role of task sequencing in mitigating CF and propose a method for determining the optimal task order. The proposed method leverages zero-shot scoring algorithms inspired by neural architecture search (NAS). Results demonstrate that intelligent task sequencing can substantially reduce CF. Moreover, when combined with traditional continual learning strategies, sequencing offers enhanced performance and robustness against forgetting. Additionally, the presented approaches can find applications in other fields, such as curriculum learning.