Efficient Personalized Adaptation for Physiological Signal Foundation Model

Chenrui Wu, Haishuai Wang, Xiang Zhang, Chengqi Zhang, Jiajun Bu
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:67833-67851, 2025.

Abstract

Time series analysis is crucial across various fields like energy, environment, transportation, finance and health. Deep learning has significantly advanced this field, particularly, the Time Series Foundation Model (TSFM) excels in multiple domains due to extensive pre-training. In this work, we focus on TSFM’s challenges in medical practice: limited computing resources and medical data privacy. TSFM variants include fine-tuned models and those pre-trained for rapid deployment on diverse data. There may not be enough computing resources to train physiological signals locally in hospitals, and generalized TSFM is still inferior to task-specific methods on private, imbalanced local data. To address this, we propose PhysioPFM, a framework for efficiently personalizing TSFM. Our approach involves low-rank pre-training on public datasets, generator training by trained LoRA weights, and efficient weight generation via local data. Experimental results demonstrate that integrating generated models with TSFM enhances performance, and transferability, and reduces the need for additional sensitive data training.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-wu25ah, title = {Efficient Personalized Adaptation for Physiological Signal Foundation Model}, author = {Wu, Chenrui and Wang, Haishuai and Zhang, Xiang and Zhang, Chengqi and Bu, Jiajun}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {67833--67851}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/wu25ah/wu25ah.pdf}, url = {https://proceedings.mlr.press/v267/wu25ah.html}, abstract = {Time series analysis is crucial across various fields like energy, environment, transportation, finance and health. Deep learning has significantly advanced this field, particularly, the Time Series Foundation Model (TSFM) excels in multiple domains due to extensive pre-training. In this work, we focus on TSFM’s challenges in medical practice: limited computing resources and medical data privacy. TSFM variants include fine-tuned models and those pre-trained for rapid deployment on diverse data. There may not be enough computing resources to train physiological signals locally in hospitals, and generalized TSFM is still inferior to task-specific methods on private, imbalanced local data. To address this, we propose PhysioPFM, a framework for efficiently personalizing TSFM. Our approach involves low-rank pre-training on public datasets, generator training by trained LoRA weights, and efficient weight generation via local data. Experimental results demonstrate that integrating generated models with TSFM enhances performance, and transferability, and reduces the need for additional sensitive data training.} }
Endnote
%0 Conference Paper %T Efficient Personalized Adaptation for Physiological Signal Foundation Model %A Chenrui Wu %A Haishuai Wang %A Xiang Zhang %A Chengqi Zhang %A Jiajun Bu %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-wu25ah %I PMLR %P 67833--67851 %U https://proceedings.mlr.press/v267/wu25ah.html %V 267 %X Time series analysis is crucial across various fields like energy, environment, transportation, finance and health. Deep learning has significantly advanced this field, particularly, the Time Series Foundation Model (TSFM) excels in multiple domains due to extensive pre-training. In this work, we focus on TSFM’s challenges in medical practice: limited computing resources and medical data privacy. TSFM variants include fine-tuned models and those pre-trained for rapid deployment on diverse data. There may not be enough computing resources to train physiological signals locally in hospitals, and generalized TSFM is still inferior to task-specific methods on private, imbalanced local data. To address this, we propose PhysioPFM, a framework for efficiently personalizing TSFM. Our approach involves low-rank pre-training on public datasets, generator training by trained LoRA weights, and efficient weight generation via local data. Experimental results demonstrate that integrating generated models with TSFM enhances performance, and transferability, and reduces the need for additional sensitive data training.
APA
Wu, C., Wang, H., Zhang, X., Zhang, C. & Bu, J.. (2025). Efficient Personalized Adaptation for Physiological Signal Foundation Model. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:67833-67851 Available from https://proceedings.mlr.press/v267/wu25ah.html.

Related Material