PASS++: A Dual Bias Reduction Framework for Non-Exemplar Class-Incremental Learning

TL;DR

PASS++ introduces a dual bias reduction framework for non-exemplar class-incremental learning, effectively mitigating representation and classifier biases without storing old data, thus reducing catastrophic forgetting.

Contribution

It proposes a novel dual bias reduction framework using self-supervised transformation and prototype augmentation, enhancing non-exemplar CIL performance.

Findings

Achieves comparable results to exemplar-based methods without storing old data.

Effectively reduces catastrophic forgetting in non-exemplar CIL.

Improves decision boundary maintenance through prototype augmentation.

Abstract

Class-incremental learning (CIL) aims to recognize new classes incrementally while maintaining the discriminability of old classes. Most existing CIL methods are exemplar-based, i.e., storing a part of old data for retraining. Without relearning old data, those methods suffer from catastrophic forgetting. In this paper, we figure out two inherent problems in CIL, i.e., representation bias and classifier bias, that cause catastrophic forgetting of old knowledge. To address these two biases, we present a simple and novel dual bias reduction framework that employs self-supervised transformation (SST) in input space and prototype augmentation (protoAug) in deep feature space. On the one hand, SST alleviates the representation bias by learning generic and diverse representations that can transfer across different tasks. On the other hand, protoAug overcomes the classifier bias by explicitly or implicitly augmenting prototypes of old classes in the deep feature space, which poses tighter constraints to maintain previously learned decision boundaries. We further propose hardness-aware prototype augmentation and multi-view ensemble strategies, leading to significant improvements. The proposed framework can be easily integrated with pre-trained models. Without storing any samples of old classes, our method can perform comparably with state-of-the-art exemplar-based approaches which store plenty of old data. We hope to draw the attention of researchers back to non-exemplar CIL by rethinking the necessity of storing old samples in CIL.