Prasanna, B and Sanyal, S and Babu, RV (2023) Continual Domain Adaptation through Pruning-aided Domain-specific Weight Modulation. In: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2023, 18 - 22 June 2023, Vancouver, BC, Canada, pp. 2457-2463.
|
PDF
IEEE-CVF_CVPRW2023_2023_2457-2463_2023.pdf - Published Version Download (1MB) | Preview |
Abstract
In this paper, we propose to develop a method to address unsupervised domain adaptation (UDA) in a practical setting of continual learning (CL). The goal is to update the model on continually changing domains while preserving domain-specific knowledge to prevent catastrophic forgetting of past-seen domains. To this end, we build a framework for preserving domain-specific features utilizing the inherent model capacity via pruning. We also perform effective inference using a novel batch-norm based metric to predict the final model parameters to be used accurately. Our approach achieves not only state-of-the-art performance but also prevents catastrophic forgetting of past domains significantly. Our code is made publicly available. 1. © 2023 IEEE.
| Item Type: | Conference Paper |
|---|---|
| Publication: | IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops |
| Publisher: | IEEE Computer Society |
| Additional Information: | The copyright for this article belongs to the Authors. |
| Keywords: | Catastrophic forgetting; Continual learning; Domain adaptation; Domain specific; Domain-specific knowledge; Modeling parameters; State-of-the-art performance, Domain Knowledge |
| Department/Centre: | Division of Interdisciplinary Sciences > Computational and Data Sciences |
| Date Deposited: | 24 Nov 2023 10:32 |
| Last Modified: | 24 Nov 2023 10:32 |
| URI: | https://eprints.iisc.ac.in/id/eprint/83213 |
Actions (login required)
 |
View Item |
