Continual Learning with Recursive Gradient Optimization
This addresses the challenge of enabling neural networks to learn multiple tasks sequentially without forgetting, which is crucial for AI systems that need to adapt over time, though it is an incremental improvement over existing methods.
The paper tackles the problem of catastrophic forgetting in continual learning by introducing Recursive Gradient Optimization (RGO), which modifies gradients to minimize forgetting without data replay, achieving state-of-the-art performance with 82.22% on 20-split-CIFAR100 and 72.63% on 20-split-miniImageNet.
Learning multiple tasks sequentially without forgetting previous knowledge, called Continual Learning(CL), remains a long-standing challenge for neural networks. Most existing methods rely on additional network capacity or data replay. In contrast, we introduce a novel approach which we refer to as Recursive Gradient Optimization(RGO). RGO is composed of an iteratively updated optimizer that modifies the gradient to minimize forgetting without data replay and a virtual Feature Encoding Layer(FEL) that represents different long-term structures with only task descriptors. Experiments demonstrate that RGO has significantly better performance on popular continual classification benchmarks when compared to the baselines and achieves new state-of-the-art performance on 20-split-CIFAR100(82.22%) and 20-split-miniImageNet(72.63%). With higher average accuracy than Single-Task Learning(STL), this method is flexible and reliable to provide continual learning capabilities for learning models that rely on gradient descent.