Novel Class Discovery without Forgetting

TL;DR

This paper introduces a new problem setting called NCDwF, focusing on discovering novel classes without forgetting previously learned categories, and proposes a method to balance knowledge retention with novel class discovery.

Contribution

The paper formulates NCDwF, proposes a novel method with pseudo-latent representations and mutual-information regularization, and establishes experimental protocols for evaluation.

Findings

Existing models catastrophically forget previous categories.

The proposed method effectively balances forgetting and novel class discovery.

Experimental results on CIFAR and ImageNet datasets demonstrate improved performance.

Abstract

Humans possess an innate ability to identify and differentiate instances that they are not familiar with, by leveraging and adapting the knowledge that they have acquired so far. Importantly, they achieve this without deteriorating the performance on their earlier learning. Inspired by this, we identify and formulate a new, pragmatic problem setting of NCDwF: Novel Class Discovery without Forgetting, which tasks a machine learning model to incrementally discover novel categories of instances from unlabeled data, while maintaining its performance on the previously seen categories. We propose 1) a method to generate pseudo-latent representations which act as a proxy for (no longer available) labeled data, thereby alleviating forgetting, 2) a mutual-information based regularizer which enhances unsupervised discovery of novel classes, and 3) a simple Known Class Identifier which aids generalized inference when the testing data contains instances form both seen and unseen categories. We introduce experimental protocols based on CIFAR-10, CIFAR-100 and ImageNet-1000 to measure the trade-off between knowledge retention and novel class discovery. Our extensive evaluations reveal that existing models catastrophically forget previously seen categories while identifying novel categories, while our method is able to effectively balance between the competing objectives. We hope our work will attract further research into this newly identified pragmatic problem setting.