Joseph, KJ and Paul, S and Aggarwal, G and Biswas, S and Rai, P and Han, K and Balasubramanian, VN (2022) Novel Class Discovery Without Forgetting. In: 17th European Conference on Computer Vision, ECCV 2022, 23 - 27 October 2022, Tel Aviv, pp. 570-586.
|
PDF
ECCV_2022.pdf - Published Version Download (3MB) | Preview |
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. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
| Item Type: | Conference Paper |
|---|---|
| Publication: | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
| Publisher: | Springer Science and Business Media Deutschland GmbH |
| Additional Information: | The copyright for this article belongs to the Author(S). |
| Keywords: | Knowledge management; Machine learning, Catastrophic forgetting; Early learning; Generalized inference; Labeled data; Machine learning models; Novel class discovery; Performance; Pseudo-latent generation and replay; Regularizer; Unlabeled data, Economic and social effects |
| Department/Centre: | Division of Electrical Sciences > Electrical Engineering |
| Date Deposited: | 09 Jan 2023 09:48 |
| Last Modified: | 09 Jan 2023 09:48 |
| URI: | https://eprints.iisc.ac.in/id/eprint/78928 |
Actions (login required)
 |
View Item |
