Robust Domain-Free Domain Generalization with Class-Aware Alignment

Page view(s)
25
Checked on Jun 16, 2024
Robust Domain-Free Domain Generalization with Class-Aware Alignment
Title:
Robust Domain-Free Domain Generalization with Class-Aware Alignment
Journal Title:
ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Publication Date:
13 May 2021
Citation:
Zhang, W., Ragab, M., & Sagarna, R. (2021). Robust Domain-Free Domain Generalization with Class-Aware Alignment. ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). doi:10.1109/icassp39728.2021.9413872
Abstract:
While deep neural networks demonstrate state-of-the-art performance on a variety of learning tasks, their performance relies on the assumption that train and test distributions are the same, which may not hold in real-world applications. Domain generalization addresses this issue by employing multiple source domains to build robust models that can generalize to unseen target domains subject to shifts in data distribution. In this paper, we propose Domain-Free Domain Generalization (DFDG), a model-agnostic method to achieve better generalization performance on the unseen test domain without the need for source domain labels. DFDG uses novel strategies to learn domain-invariant class-discriminative features. It aligns class relationships of samples through class-conditional soft labels, and uses saliency maps, traditionally developed for post-hoc analysis of image classification networks, to remove superficial observations from training inputs. DFDG obtains competitive performance on both time series sensor and image classification public datasets.
License type:
Publisher Copyright
Funding Info:
There was no specific funding for the research done
Description:
© 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
ISSN:
2379-190X
1520-6149
ISBN:
978-1-7281-7605-5
978-1-7281-7606-2
Files uploaded:

File Size Format Action
domain-generalization-icassp-final.pdf 966.88 KB PDF Open