Reducing Catastrophic Forgetting in Online Class Incremental Learning Using Self-Distillation

TL;DR

This paper proposes a self-distillation approach to mitigate catastrophic forgetting in online class incremental learning, enhancing knowledge transfer and memory management for better performance on standard datasets.

Contribution

It introduces a novel self-distillation technique using a highly generalizable shallow layer output as a teacher, and a memory update strategy prioritizing misclassified samples.

Findings

Outperforms conventional methods on CIFAR10, CIFAR100, MiniImageNet

Improves knowledge transfer and memory efficiency

Reduces catastrophic forgetting in incremental learning

Abstract

In continual learning, there is a serious problem of catastrophic forgetting, in which previous knowledge is forgotten when a model learns new tasks. Various methods have been proposed to solve this problem. Replay methods which replay data from previous tasks in later training, have shown good accuracy. However, replay methods have a generalizability problem from a limited memory buffer. In this paper, we tried to solve this problem by acquiring transferable knowledge through self-distillation using highly generalizable output in shallow layer as a teacher. Furthermore, when we deal with a large number of classes or challenging data, there is a risk of learning not converging and not experiencing overfitting. Therefore, we attempted to achieve more efficient and thorough learning by prioritizing the storage of easily misclassified samples through a new method of memory update. We confirmed that our proposed method outperformed conventional methods by experiments on CIFAR10, CIFAR100, and MiniimageNet datasets.