Accelerated Methods with Complexity Separation Under Data Similarity for Federated Learning Problems
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914251897044992 |
|---|---|
| author | Bylinkin, Dmitry Skorik, Sergey Bystrov, Dmitriy Berezin, Leonid Avetisyan, Aram Beznosikov, Aleksandr |
| author_facet | Bylinkin, Dmitry Skorik, Sergey Bystrov, Dmitriy Berezin, Leonid Avetisyan, Aram Beznosikov, Aleksandr |
| contents | Heterogeneity within data distribution poses a challenge in many modern federated learning tasks. We formalize it as an optimization problem involving a computationally heavy composite under data similarity. By employing different sets of assumptions, we present several approaches to develop communication-efficient methods. An optimal algorithm is proposed for the convex case. The constructed theory is validated through a series of experiments across various problems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_08614 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Accelerated Methods with Complexity Separation Under Data Similarity for Federated Learning Problems Bylinkin, Dmitry Skorik, Sergey Bystrov, Dmitriy Berezin, Leonid Avetisyan, Aram Beznosikov, Aleksandr Optimization and Control Machine Learning Heterogeneity within data distribution poses a challenge in many modern federated learning tasks. We formalize it as an optimization problem involving a computationally heavy composite under data similarity. By employing different sets of assumptions, we present several approaches to develop communication-efficient methods. An optimal algorithm is proposed for the convex case. The constructed theory is validated through a series of experiments across various problems. |
| title | Accelerated Methods with Complexity Separation Under Data Similarity for Federated Learning Problems |
| topic | Optimization and Control Machine Learning |
| url | https://arxiv.org/abs/2601.08614 |