Better Knowledge Retention through Metric Learning

TL;DR

This paper introduces a novel metric learning approach for continual learning that significantly reduces forgetting in deep neural networks when new categories are introduced, outperforming existing methods on CIFAR-10 and ImageNet.

Contribution

The paper presents a new metric learning method that enhances knowledge retention and mitigates forgetting in deep neural networks during continual learning.

Findings

Reduces forgetting by 2.3x to 6.9x on CIFAR-10

Reduces forgetting by 1.8x to 2.7x on ImageNet

Outperforms existing methods and approaches an oracle baseline

Abstract

In continual learning, new categories may be introduced over time, and an ideal learning system should perform well on both the original categories and the new categories. While deep neural nets have achieved resounding success in the classical supervised setting, they are known to forget about knowledge acquired in prior episodes of learning if the examples encountered in the current episode of learning are drastically different from those encountered in prior episodes. In this paper, we propose a new method that can both leverage the expressive power of deep neural nets and is resilient to forgetting when new categories are introduced. We found the proposed method can reduce forgetting by 2.3x to 6.9x on CIFAR-10 compared to existing methods and by 1.8x to 2.7x on ImageNet compared to an oracle baseline.