Information-Geometric Barycenters for Bayesian Federated Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911184309977088 |
|---|---|
| author | Jamoussi, Nour Serra, Giuseppe Stavrou, Photios A. Kountouris, Marios |
| author_facet | Jamoussi, Nour Serra, Giuseppe Stavrou, Photios A. Kountouris, Marios |
| contents | Federated learning (FL) is a widely used and impactful distributed optimization framework that achieves consensus through averaging locally trained models. While effective, this approach may not align well with Bayesian inference, where the model space has the structure of a distribution space. Taking an information-geometric perspective, we reinterpret FL aggregation as the problem of finding the barycenter of local posteriors using a prespecified divergence metric, minimizing the average discrepancy across clients. This perspective provides a unifying framework that generalizes many existing methods and offers crisp insights into their theoretical underpinnings. We then propose BA-BFL, an algorithm that retains the convergence properties of Federated Averaging in non-convex settings. In non-independent and identically distributed scenarios, we conduct extensive comparisons with statistical aggregation techniques, showing that BA-BFL achieves performance comparable to state-of-the-art methods while offering a geometric interpretation of the aggregation phase. Additionally, we extend our analysis to Hybrid Bayesian Deep Learning, exploring the impact of Bayesian layers on uncertainty quantification and model calibration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_11646 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Information-Geometric Barycenters for Bayesian Federated Learning Jamoussi, Nour Serra, Giuseppe Stavrou, Photios A. Kountouris, Marios Machine Learning Information Theory Networking and Internet Architecture Federated learning (FL) is a widely used and impactful distributed optimization framework that achieves consensus through averaging locally trained models. While effective, this approach may not align well with Bayesian inference, where the model space has the structure of a distribution space. Taking an information-geometric perspective, we reinterpret FL aggregation as the problem of finding the barycenter of local posteriors using a prespecified divergence metric, minimizing the average discrepancy across clients. This perspective provides a unifying framework that generalizes many existing methods and offers crisp insights into their theoretical underpinnings. We then propose BA-BFL, an algorithm that retains the convergence properties of Federated Averaging in non-convex settings. In non-independent and identically distributed scenarios, we conduct extensive comparisons with statistical aggregation techniques, showing that BA-BFL achieves performance comparable to state-of-the-art methods while offering a geometric interpretation of the aggregation phase. Additionally, we extend our analysis to Hybrid Bayesian Deep Learning, exploring the impact of Bayesian layers on uncertainty quantification and model calibration. |
| title | Information-Geometric Barycenters for Bayesian Federated Learning |
| topic | Machine Learning Information Theory Networking and Internet Architecture |
| url | https://arxiv.org/abs/2412.11646 |