Achieving Upper Bound Accuracy of Joint Training in Continual Learning
This solves the accuracy gap problem for continual learning applications, making it ready for real-life use, though it appears incremental as it builds on existing theory and models.
The paper addresses the gap between state-of-the-art continual learning algorithms and the ideal upper-bound accuracy achieved by joint training, and reports that leveraging theory and large foundation models has enabled achieving this upper-bound accuracy, as validated empirically on text and image datasets.
Continual learning has been an active research area in machine learning, focusing on incrementally learning a sequence of tasks. A key challenge is catastrophic forgetting (CF), and most research efforts have been directed toward mitigating this issue. However, a significant gap remains between the accuracy achieved by state-of-the-art continual learning algorithms and the ideal or upper-bound accuracy achieved by training all tasks together jointly. This gap has hindered or even prevented the adoption of continual learning in applications, as accuracy is often of paramount importance. Recently, another challenge, termed inter-task class separation (ICS), was also identified, which spurred a theoretical study into principled approaches for solving continual learning. Further research has shown that by leveraging the theory and the power of large foundation models, it is now possible to achieve upper-bound accuracy, which has been empirically validated using both text and image classification datasets. Continual learning is now ready for real-life applications. This paper surveys the main research leading to this achievement, justifies the approach both intuitively and from neuroscience research, and discusses insights gained.