AggSS: An Aggregated Self-Supervised Approach for Class-Incremental Learning
This work addresses the problem of catastrophic forgetting in class-incremental learning for machine learning practitioners, offering a plug-and-play module that is incremental in nature.
The paper tackles class-incremental learning by proposing AggSS, an aggregated self-supervised approach using image rotations to enhance feature learning, resulting in significant performance improvements on CIFAR-100 and ImageNet-Subset datasets.
This paper investigates the impact of self-supervised learning, specifically image rotations, on various class-incremental learning paradigms. Here, each image with a predefined rotation is considered as a new class for training. At inference, all image rotation predictions are aggregated for the final prediction, a strategy we term Aggregated Self-Supervision (AggSS). We observe a shift in the deep neural network's attention towards intrinsic object features as it learns through AggSS strategy. This learning approach significantly enhances class-incremental learning by promoting robust feature learning. AggSS serves as a plug-and-play module that can be seamlessly incorporated into any class-incremental learning framework, leveraging its powerful feature learning capabilities to enhance performance across various class-incremental learning approaches. Extensive experiments conducted on standard incremental learning datasets CIFAR-100 and ImageNet-Subset demonstrate the significant role of AggSS in improving performance within these paradigms.