Robust Federated Personalised Mean Estimation for the Gaussian Mixture Model
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908443885961216 |
|---|---|
| author | Managoli, Malhar A. Prabhakaran, Vinod M. Diggavi, Suhas |
| author_facet | Managoli, Malhar A. Prabhakaran, Vinod M. Diggavi, Suhas |
| contents | Federated learning with heterogeneous data and personalization has received significant recent attention. Separately, robustness to corrupted data in the context of federated learning has also been studied. In this paper we explore combining personalization for heterogeneous data with robustness, where a constant fraction of the clients are corrupted. Motivated by this broad problem, we formulate a simple instantiation which captures some of its difficulty. We focus on the specific problem of personalized mean estimation where the data is drawn from a Gaussian mixture model. We give an algorithm whose error depends almost linearly on the ratio of corrupted to uncorrupted samples, and show a lower bound with the same behavior, albeit with a gap of a constant factor. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_19955 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Robust Federated Personalised Mean Estimation for the Gaussian Mixture Model Managoli, Malhar A. Prabhakaran, Vinod M. Diggavi, Suhas Machine Learning Information Theory Federated learning with heterogeneous data and personalization has received significant recent attention. Separately, robustness to corrupted data in the context of federated learning has also been studied. In this paper we explore combining personalization for heterogeneous data with robustness, where a constant fraction of the clients are corrupted. Motivated by this broad problem, we formulate a simple instantiation which captures some of its difficulty. We focus on the specific problem of personalized mean estimation where the data is drawn from a Gaussian mixture model. We give an algorithm whose error depends almost linearly on the ratio of corrupted to uncorrupted samples, and show a lower bound with the same behavior, albeit with a gap of a constant factor. |
| title | Robust Federated Personalised Mean Estimation for the Gaussian Mixture Model |
| topic | Machine Learning Information Theory |
| url | https://arxiv.org/abs/2504.19955 |