ACAE-REMIND for Online Continual Learning with Compressed Feature Replay
This work addresses the challenge of efficient and effective continual learning for AI systems that process streaming data, representing an incremental improvement over existing feature replay methods.
The paper tackled the problem of limited representation learning in online continual learning due to fixed backbone networks by proposing an auxiliary classifier auto-encoder module for compressed feature replay, achieving state-of-the-art performance on ImageNet-Subset, CIFAR100, and CIFAR10 datasets.
Online continual learning aims to learn from a non-IID stream of data from a number of different tasks, where the learner is only allowed to consider data once. Methods are typically allowed to use a limited buffer to store some of the images in the stream. Recently, it was found that feature replay, where an intermediate layer representation of the image is stored (or generated) leads to superior results than image replay, while requiring less memory. Quantized exemplars can further reduce the memory usage. However, a drawback of these methods is that they use a fixed (or very intransigent) backbone network. This significantly limits the learning of representations that can discriminate between all tasks. To address this problem, we propose an auxiliary classifier auto-encoder (ACAE) module for feature replay at intermediate layers with high compression rates. The reduced memory footprint per image allows us to save more exemplars for replay. In our experiments, we conduct task-agnostic evaluation under online continual learning setting and get state-of-the-art performance on ImageNet-Subset, CIFAR100 and CIFAR10 dataset.