Continual Learning by Asymmetric Loss Approximation with Single-Side Overestimation
This addresses the problem of catastrophic forgetting for AI systems requiring scalable and efficient continual learning without additional network components.
The paper tackles catastrophic forgetting in deep neural networks by proposing a continual learning method that approximates the true loss function with an asymmetric quadratic function, achieving state-of-the-art accuracy close to upper-bound performance on challenging benchmarks.
Catastrophic forgetting is a critical challenge in training deep neural networks. Although continual learning has been investigated as a countermeasure to the problem, it often suffers from the requirements of additional network components and the limited scalability to a large number of tasks. We propose a novel approach to continual learning by approximating a true loss function using an asymmetric quadratic function with one of its sides overestimated. Our algorithm is motivated by the empirical observation that the network parameter updates affect the target loss functions asymmetrically. In the proposed continual learning framework, we estimate an asymmetric loss function for the tasks considered in the past through a proper overestimation of its unobserved sides in training new tasks, while deriving the accurate model parameter for the observable sides. In contrast to existing approaches, our method is free from the side effects and achieves the state-of-the-art accuracy that is even close to the upper-bound performance on several challenging benchmark datasets.