Please use this identifier to cite or link to this item:
https://hdl.handle.net/10356/182272
Title: | ContrastSense: domain-invariant contrastive learning for in-the-wild wearable sensing | Authors: | Dai, Gaole Xu, Huatao Yoon, Hyungjun Li, Mo Tan, Rui Lee, Sung-Ju |
Keywords: | Computer and Information Science | Issue Date: | 2024 | Source: | Dai, G., Xu, H., Yoon, H., Li, M., Tan, R. & Lee, S. (2024). ContrastSense: domain-invariant contrastive learning for in-the-wild wearable sensing. Proceedings of the ACM On Interactive, Mobile, Wearable and Ubiquitous Technologies, 8(4), 3699744-. https://dx.doi.org/10.1145/3699744 | Project: | MOE-T2EP20220-0004 | Journal: | Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies | Abstract: | Existing wearable sensing models often struggle with domain shifts and class label scarcity. Contrastive learning is a promising technique to address class label scarcity, which however captures domain-related features and suffers from low-quality negatives. To address both problems, we propose ContrastSense, a domain-invariant contrastive learning scheme for a realistic wearable sensing scenario where domain shifts and class label scarcity are presented simultaneously. To capture domain-invariant information, ContrastSense exploits unlabeled data and domain labels specifying user IDs or devices to minimize the discrepancy across domains. To improve the quality of negatives, time and domain labels are leveraged to select samples and refine negatives. In addition, ContrastSense designs a parameter-wise penalty to preserve domaininvariant knowledge during fine-tuning to further maintain model robustness. Extensive experiments show that ContrastSense outperforms the state-of-the-art baselines by 8.9% on human activity recognition with inertial measurement units and 5.6% on gesture recognition with electromyography when presented with domain shifts across users. Besides, when presented with different kinds of domain shifts across devices, on-body positions, and datasets, ContrastSense achieves consistent improvements compared with the best baselines. | URI: | https://hdl.handle.net/10356/182272 | ISSN: | 2474-9567 | DOI: | 10.1145/3699744 | Schools: | College of Computing and Data Science | Rights: | © 2024 the Owner/Author(s). This work is licensed under a Creative Commons Attribution International 4.0 License. | Fulltext Permission: | open | Fulltext Availability: | With Fulltext |
Appears in Collections: | CCDS Journal Articles |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
3699744.pdf | 1.43 MB | Adobe PDF | View/Open |
Page view(s)
74
Updated on Feb 16, 2025
Download(s)
10
Updated on Feb 16, 2025
Google ScholarTM
Check
Altmetric
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.