Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2510.03726 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916990160994304 |
|---|---|
| author | Zeng, Jiahao Xing, Wolong Shi, Liangtao Huang, Xin Wang, Jialin Cao, Zhile Shi, Zhenkui |
| author_facet | Zeng, Jiahao Xing, Wolong Shi, Liangtao Huang, Xin Wang, Jialin Cao, Zhile Shi, Zhenkui |
| contents | Federated learning has received significant attention for its ability to simultaneously protect customer privacy and leverage distributed data from multiple devices for model training. However, conventional approaches often focus on isolated heterogeneous scenarios, resulting in skewed feature distributions or label distributions. Meanwhile, data heterogeneity is actually a key factor in improving model performance. To address this issue, we propose a new approach called PFPL in mixed heterogeneous scenarios. The method provides richer domain knowledge and unbiased convergence targets by constructing personalized, unbiased prototypes for each client. Moreover, in the local update phase, we introduce consistent regularization to align local instances with their personalized prototypes, which significantly improves the convergence of the loss function. Experimental results on Digits and Office Caltech datasets validate the effectiveness of our approach and successfully reduce the communication cost. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_03726 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Personalized federated prototype learning in mixed heterogeneous data scenarios Zeng, Jiahao Xing, Wolong Shi, Liangtao Huang, Xin Wang, Jialin Cao, Zhile Shi, Zhenkui Machine Learning Federated learning has received significant attention for its ability to simultaneously protect customer privacy and leverage distributed data from multiple devices for model training. However, conventional approaches often focus on isolated heterogeneous scenarios, resulting in skewed feature distributions or label distributions. Meanwhile, data heterogeneity is actually a key factor in improving model performance. To address this issue, we propose a new approach called PFPL in mixed heterogeneous scenarios. The method provides richer domain knowledge and unbiased convergence targets by constructing personalized, unbiased prototypes for each client. Moreover, in the local update phase, we introduce consistent regularization to align local instances with their personalized prototypes, which significantly improves the convergence of the loss function. Experimental results on Digits and Office Caltech datasets validate the effectiveness of our approach and successfully reduce the communication cost. |
| title | Personalized federated prototype learning in mixed heterogeneous data scenarios |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2510.03726 |