BSemiFL: Semi-supervised Federated Learning via a Bayesian Approach

Haozhao Wang, Shengyu Wang, Jiaming Li, Hao Ren, Xingshuo Han, Wenchao Xu, Shangwei Guo, Tianwei Zhang, Ruixuan Li
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:64655-64669, 2025.

Abstract

Semi-supervised Federated Learning (SSFL) is a promising approach that allows clients to collaboratively train a global model in the absence of their local data labels. The key step of SSFL is the re-labeling where each client adopts two types of available models, namely global and local models, to re-label the local data. While various technologies such as using the global model or the average of two models have been proposed to conduct the re-labeling step, little literature delves deeply into the performance dominance and limitations of the two models. In this paper, we first theoretically and empirically demonstrate that the local model achieves higher re-labeling accuracy over local data while the global model can progressively improve the re-labeling performance by introducing the extra data knowledge of other clients. Based on these findings, we propose BSemiFL which re-labels the local data through the collaboration between the local and global model in a Bayesian approach. Specifically, to re-label any given local sample, BSemiFL first uses Bayesian inference to assess the closeness of the local/global model to the sample. Then, it applies a weighted combination of their pseudo labels, using the closeness as the weights. Theoretical analysis shows that the labeling error of our method is smaller than that of simply using the global model, the local model, or their simple average. Experimental results show that BSemiFL improves the performance by up to $9.8%$ as compared to state-of-the-art methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-wang25dc, title = {{BS}emi{FL}: Semi-supervised Federated Learning via a {B}ayesian Approach}, author = {Wang, Haozhao and Wang, Shengyu and Li, Jiaming and Ren, Hao and Han, Xingshuo and Xu, Wenchao and Guo, Shangwei and Zhang, Tianwei and Li, Ruixuan}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {64655--64669}, 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/wang25dc/wang25dc.pdf}, url = {https://proceedings.mlr.press/v267/wang25dc.html}, abstract = {Semi-supervised Federated Learning (SSFL) is a promising approach that allows clients to collaboratively train a global model in the absence of their local data labels. The key step of SSFL is the re-labeling where each client adopts two types of available models, namely global and local models, to re-label the local data. While various technologies such as using the global model or the average of two models have been proposed to conduct the re-labeling step, little literature delves deeply into the performance dominance and limitations of the two models. In this paper, we first theoretically and empirically demonstrate that the local model achieves higher re-labeling accuracy over local data while the global model can progressively improve the re-labeling performance by introducing the extra data knowledge of other clients. Based on these findings, we propose BSemiFL which re-labels the local data through the collaboration between the local and global model in a Bayesian approach. Specifically, to re-label any given local sample, BSemiFL first uses Bayesian inference to assess the closeness of the local/global model to the sample. Then, it applies a weighted combination of their pseudo labels, using the closeness as the weights. Theoretical analysis shows that the labeling error of our method is smaller than that of simply using the global model, the local model, or their simple average. Experimental results show that BSemiFL improves the performance by up to $9.8%$ as compared to state-of-the-art methods.} }
Endnote
%0 Conference Paper %T BSemiFL: Semi-supervised Federated Learning via a Bayesian Approach %A Haozhao Wang %A Shengyu Wang %A Jiaming Li %A Hao Ren %A Xingshuo Han %A Wenchao Xu %A Shangwei Guo %A Tianwei Zhang %A Ruixuan Li %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-wang25dc %I PMLR %P 64655--64669 %U https://proceedings.mlr.press/v267/wang25dc.html %V 267 %X Semi-supervised Federated Learning (SSFL) is a promising approach that allows clients to collaboratively train a global model in the absence of their local data labels. The key step of SSFL is the re-labeling where each client adopts two types of available models, namely global and local models, to re-label the local data. While various technologies such as using the global model or the average of two models have been proposed to conduct the re-labeling step, little literature delves deeply into the performance dominance and limitations of the two models. In this paper, we first theoretically and empirically demonstrate that the local model achieves higher re-labeling accuracy over local data while the global model can progressively improve the re-labeling performance by introducing the extra data knowledge of other clients. Based on these findings, we propose BSemiFL which re-labels the local data through the collaboration between the local and global model in a Bayesian approach. Specifically, to re-label any given local sample, BSemiFL first uses Bayesian inference to assess the closeness of the local/global model to the sample. Then, it applies a weighted combination of their pseudo labels, using the closeness as the weights. Theoretical analysis shows that the labeling error of our method is smaller than that of simply using the global model, the local model, or their simple average. Experimental results show that BSemiFL improves the performance by up to $9.8%$ as compared to state-of-the-art methods.
APA
Wang, H., Wang, S., Li, J., Ren, H., Han, X., Xu, W., Guo, S., Zhang, T. & Li, R.. (2025). BSemiFL: Semi-supervised Federated Learning via a Bayesian Approach. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:64655-64669 Available from https://proceedings.mlr.press/v267/wang25dc.html.

Related Material