{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,17]],"date-time":"2026-03-17T08:00:46Z","timestamp":1773734446100,"version":"3.50.1"},"reference-count":61,"publisher":"Association for Computing Machinery (ACM)","issue":"1","license":[{"start":{"date-parts":[[2025,3,3]],"date-time":"2025-03-03T00:00:00Z","timestamp":1740960000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"name":"RGC under Contract CERG","award":["16206122, 16204523, AoE\/E-601\/22-R, R6021-20, and Contract R8015"],"award-info":[{"award-number":["16206122, 16204523, AoE\/E-601\/22-R, R6021-20, and Contract R8015"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["Proc. ACM Interact. Mob. Wearable Ubiquitous Technol."],"published-print":{"date-parts":[[2025,3,3]]},"abstract":"<jats:p>Human Activity Recognition (HAR) faces significant challenges when deployed in real-world scenarios due to non-independent and identically distributed (non-IID) data distributions. While existing domain adaptation (DA) approaches attempt to address this issue, they either require access to source data or struggle with large domain shifts. This paper presents a novel source-free domain adaptation framework for HAR that effectively handles substantial domain discrepancies across different datasets. Our approach introduces two key innovations: (1) a Discriminative Information Gramian (DIG) method that quantifies the relationship between target-domain samples and the source domain without requiring access to source data, and (2) an unsupervised domain generalization technique that ensures consistent feature extraction across augmented data samples, enhancing the model's effectiveness in the target domain. We evaluate our framework on five diverse HAR datasets comprising 87 users with varying demographics, devices, and environmental conditions. In single-source scenarios, our method achieves 76.77% accuracy and 67.03% F1-score, surpassing state-of-the-art solutions by 9.77% and 17.43%, respectively. For multi-source scenarios, we attain 85.33% accuracy and 79.55% F1-score, exceeding existing methods by at least 8.8% and 14.8%, respectively. This work represents the first successful attempt at dataset-level HAR domain adaptation without access to source data, marking a significant advancement in practical HAR applications.<\/jats:p>","DOI":"10.1145\/3712620","type":"journal-article","created":{"date-parts":[[2025,3,4]],"date-time":"2025-03-04T12:10:14Z","timestamp":1741090214000},"page":"1-24","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":3,"title":["MobHAR: Source-free Knowledge Transfer for Human Activity Recognition on Mobile Devices"],"prefix":"10.1145","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4745-6526","authenticated-orcid":false,"given":"Meng","family":"Xue","sequence":"first","affiliation":[{"name":"The Hong Kong University of Science and Technology, Hong Kong SAR"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-8304-0844","authenticated-orcid":false,"given":"Yinan","family":"Zhu","sequence":"additional","affiliation":[{"name":"The Hong Kong University of Science and Technology, Hong Kong SAR"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7858-4419","authenticated-orcid":false,"given":"Wentao","family":"Xie","sequence":"additional","affiliation":[{"name":"The Hong Kong University of Science and Technology, Hong Kong SAR"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-1519-9314","authenticated-orcid":false,"given":"Zhixian","family":"Wang","sequence":"additional","affiliation":[{"name":"Wuhan University, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1382-0679","authenticated-orcid":false,"given":"Yanjiao","family":"Chen","sequence":"additional","affiliation":[{"name":"Zhejiang University, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4055-7503","authenticated-orcid":false,"given":"Kui","family":"Jiang","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9205-1881","authenticated-orcid":false,"given":"Qian","family":"Zhang","sequence":"additional","affiliation":[{"name":"The Hong Kong University of Science and Technology, Hong Kong SAR"}]}],"member":"320","published-online":{"date-parts":[[2025,3,4]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"International conference on machine learning. PMLR, 214--223","author":"Arjovsky Martin","year":"2017","unstructured":"Martin Arjovsky, Soumith Chintala, and L\u00e9on Bottou. 2017. Wasserstein generative adversarial networks. In International conference on machine learning. PMLR, 214--223."},{"key":"e_1_2_1_2_1","volume-title":"A theory of learning from different domains. Machine learning 79","author":"Ben-David Shai","year":"2010","unstructured":"Shai Ben-David, John Blitzer, Koby Crammer, Alex Kulesza, Fernando Pereira, and Jennifer Wortman Vaughan. 2010. A theory of learning from different domains. Machine learning 79 (2010), 151--175."},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2019.2920283"},{"key":"e_1_2_1_4_1","volume-title":"Pattern recognition and machine learning","author":"Bishop Christopher M","year":"2006","unstructured":"Christopher M Bishop. 2006. Pattern recognition and machine learning. Springer google schola 2 (2006), 5--43."},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1145\/3699776"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1145\/3380985"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00704"},{"key":"e_1_2_1_8_1","volume-title":"Fast batch nuclear-norm maximization and minimization for robust domain adaptation. arXiv preprint arXiv:2107.06154","author":"Cui Shuhao","year":"2021","unstructured":"Shuhao Cui, Shuhui Wang, Junbao Zhuo, Liang Li, Qingming Huang, and Qi Tian. 2021. Fast batch nuclear-norm maximization and minimization for robust domain adaptation. arXiv preprint arXiv:2107.06154 (2021)."},{"key":"e_1_2_1_9_1","volume-title":"Testing hypotheses by regularized maximum mean discrepancy. arXiv preprint arXiv:1305.0423","author":"Danafar Somayeh","year":"2013","unstructured":"Somayeh Danafar, Paola Rancoita, Tobias Glasmachers, Kevin Whittingstall, and Juergen Schmidhuber. 2013. Testing hypotheses by regularized maximum mean discrepancy. arXiv preprint arXiv:1305.0423 (2013)."},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01411"},{"key":"e_1_2_1_11_1","first-page":"1","article-title":"Domain-adversarial training of neural networks","volume":"17","author":"Ganin Yaroslav","year":"2016","unstructured":"Yaroslav Ganin, Evgeniya Ustinova, Hana Ajakan, Pascal Germain, Hugo Larochelle, Fran\u00e7ois Laviolette, Mario March, and Victor Lempitsky. 2016. Domain-adversarial training of neural networks. Journal of machine learning research 17, 59 (2016), 1--35.","journal-title":"Journal of machine learning research"},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.265"},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1145\/3596256"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1145\/3356250.3360020"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/3191743"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.5555\/2503308.2188410"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/3463506"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3659597"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3517246"},{"key":"e_1_2_1_20_1","volume-title":"Proceedings of the International Conference on Learning Representations (ICLR)","author":"Jang Minguk","year":"2023","unstructured":"Minguk Jang, Sae-Young Chung, and Hye Won Chung. 2023. Test-time adaptation via self-training with nearest neighbor information. Proceedings of the International Conference on Learning Representations (ICLR) (2023)."},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/2733373.2806333"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58589-1_28"},{"key":"e_1_2_1_23_1","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3631428","article-title":"SF-Adapter: Computational-Efficient Source-Free Domain Adaptation for Human Activity Recognition","volume":"7","author":"Kang Hua","year":"2024","unstructured":"Hua Kang, Qingyong Hu, and Qian Zhang. 2024. SF-Adapter: Computational-Efficient Source-Free Domain Adaptation for Human Activity Recognition. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 7, 4 (2024), 1--23.","journal-title":"Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1109\/PERCOM.2018.8444585"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jesp.2013.03.013"},{"key":"e_1_2_1_26_1","first-page":"1","article-title":"Neuralgait: Assessing brain health using your smartphone","volume":"6","author":"Li Huining","year":"2023","unstructured":"Huining Li, Huan Chen, Chenhan Xu, Zhengxiong Li, Hanbin Zhang, Xiaoye Qian, Dongmei Li, Ming-chun Huang, and Wenyao Xu. 2023. Neuralgait: Assessing brain health using your smartphone. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 6, 4 (2023), 1--28.","journal-title":"Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies"},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/3432208"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/3534589"},{"key":"e_1_2_1_29_1","volume-title":"The\" Beatrix\"Resurrections: Robust Backdoor Detection via Gram Matrices. arXiv preprint arXiv:2209.11715","author":"Ma Wanlun","year":"2022","unstructured":"Wanlun Ma, Derui Wang, Ruoxi Sun, Minhui Xue, Sheng Wen, and Yang Xiang. 2022. The\" Beatrix\"Resurrections: Robust Backdoor Detection via Gram Matrices. arXiv preprint arXiv:2209.11715 (2022)."},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1145\/3302505.3310068"},{"key":"e_1_2_1_31_1","doi-asserted-by":"crossref","unstructured":"Krikamol Muandet Kenji Fukumizu Bharath Sriperumbudur Bernhard Sch\u00f6lkopf et al. 2017. Kernel mean embedding of distributions: A review and beyond. Foundations and Trends\u00ae in Machine Learning 10 1--2 (2017) 1--141.","DOI":"10.1561\/2200000060"},{"key":"e_1_2_1_32_1","volume-title":"11th International Conference on Learning Representations","author":"Niu Shuaicheng","year":"2023","unstructured":"Shuaicheng Niu, Jiaxiang Wu, Yifan Zhang, Zhiquan Wen, Yaofo Chen, Peilin Zhao, and Mingkui Tan. 2023. Towards stable test-time adaptation in dynamic wild world. In 11th International Conference on Learning Representations (2023)."},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-020-09904-8"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i13.17416"},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/3534678.3539134"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1145\/3369818"},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/3552434"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1145\/3580305.3599360"},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2015.07.085"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3328932"},{"key":"e_1_2_1_41_1","unstructured":"Meyer Scetbon and Ga\u00ebl Varoquaux. 2019. Comparing distributions: l1 geometry improves kernel two-sample testing. In NeurIPS 2019-33th Conference on Neural Information Processing Systems."},{"key":"e_1_2_1_42_1","doi-asserted-by":"publisher","DOI":"10.3390\/s140610146"},{"key":"e_1_2_1_43_1","first-page":"14638","article-title":"Top-k training of gans: Improving gan performance by throwing away bad samples","volume":"33","author":"Sinha Samarth","year":"2020","unstructured":"Samarth Sinha, Zhengli Zhao, Anirudh Goyal ALIAS PARTH GOYAL, Colin A Raffel, and Augustus Odena. 2020. Top-k training of gans: Improving gan performance by throwing away bad samples. Advances in Neural Information Processing Systems 33 (2020), 14638--14649.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","DOI":"10.1145\/2809695.2809718"},{"key":"e_1_2_1_45_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i7.16763"},{"key":"e_1_2_1_46_1","doi-asserted-by":"publisher","DOI":"10.1145\/3448112"},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.1145\/3136755.3136817"},{"key":"e_1_2_1_48_1","article-title":"Visualizing data using t-SNE","volume":"9","author":"der Maaten Laurens Van","year":"2008","unstructured":"Laurens Van der Maaten and Geoffrey Hinton. 2008. Visualizing data using t-SNE. Journal of machine learning research 9, 11 (2008).","journal-title":"Journal of machine learning research"},{"key":"e_1_2_1_49_1","volume-title":"9th International Conference on Learning Representations","author":"Wang Dequan","year":"2021","unstructured":"Dequan Wang, Evan Shelhamer, Shaoteng Liu, Bruno Olshausen, and Trevor Darrell. 2021. Tent: Fully test-time adaptation by entropy minimization. In 9th International Conference on Learning Representations (2021)."},{"key":"e_1_2_1_50_1","volume-title":"Generalizing to unseen domains: A survey on domain generalization","author":"Wang Jindong","year":"2022","unstructured":"Jindong Wang, Cuiling Lan, Chang Liu, Yidong Ouyang, Tao Qin, Wang Lu, Yiqiang Chen, Wenjun Zeng, and Philip Yu. 2022. Generalizing to unseen domains: A survey on domain generalization. IEEE Transactions on Knowledge and Data Engineering (2022)."},{"key":"e_1_2_1_51_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSEN.2022.3214198"},{"key":"e_1_2_1_52_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00706"},{"key":"e_1_2_1_53_1","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3631450","article-title":"Optimization-Free Test-Time Adaptation for Cross-Person Activity Recognition","volume":"7","author":"Wang Shuoyuan","year":"2024","unstructured":"Shuoyuan Wang, Jindong Wang, Huajun Xi, Bob Zhang, Lei Zhang, and Hongxin Wei. 2024. Optimization-Free Test-Time Adaptation for Cross-Person Activity Recognition. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 7, 4 (2024), 1--27.","journal-title":"Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies"},{"key":"e_1_2_1_54_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2019.04.057"},{"key":"e_1_2_1_55_1","doi-asserted-by":"publisher","DOI":"10.1145\/3570361.3613299"},{"key":"e_1_2_1_56_1","doi-asserted-by":"publisher","DOI":"10.1145\/3485730.3485937"},{"key":"e_1_2_1_57_1","doi-asserted-by":"publisher","DOI":"10.1145\/3038912.3052577"},{"key":"e_1_2_1_58_1","volume-title":"International Conference on Learning Representations. https:\/\/openreview.net\/forum?id=Sy8gdB9xx","author":"Zhang Chiyuan","year":"2017","unstructured":"Chiyuan Zhang, Samy Bengio, Moritz Hardt, Benjamin Recht, and Oriol Vinyals. 2017. Understanding deep learning requires rethinking generalization. In International Conference on Learning Representations. https:\/\/openreview.net\/forum?id=Sy8gdB9xx"},{"key":"e_1_2_1_59_1","doi-asserted-by":"publisher","DOI":"10.1145\/3637528.3671828"},{"key":"e_1_2_1_60_1","volume-title":"Sawchuk","author":"Zhang Mi","year":"2012","unstructured":"Mi Zhang and Alexander A. Sawchuk. 2012. USC-HAD: A Daily Activity Dataset for Ubiquitous Activity Recognition Using Wearable Sensors. In ACM International Conference on Ubiquitous Computing (Ubicomp) Workshop on Situation, Activity and Goal Awareness (SAGAware). Pittsburgh, Pennsylvania, USA."},{"key":"e_1_2_1_61_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2022.3229621"}],"container-title":["Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3712620","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3712620","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,22]],"date-time":"2025-08-22T19:29:50Z","timestamp":1755890990000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3712620"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,3]]},"references-count":61,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2025,3,3]]}},"alternative-id":["10.1145\/3712620"],"URL":"https:\/\/doi.org\/10.1145\/3712620","relation":{},"ISSN":["2474-9567"],"issn-type":[{"value":"2474-9567","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,3]]},"assertion":[{"value":"2025-03-04","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}