Exploring New Frontiers in Vertical Federated Learning: the Role of Saddle Point Reformulation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910025497182208 |
|---|---|
| author | Beznosikov, Aleksandr Kormakov, Georgiy Grigorievskiy, Alexander Rudakov, Mikhail Nazykov, Ruslan Rogozin, Alexander Vakhrushev, Anton Savchenko, Andrey Takáč, Martin Gasnikov, Alexander |
| author_facet | Beznosikov, Aleksandr Kormakov, Georgiy Grigorievskiy, Alexander Rudakov, Mikhail Nazykov, Ruslan Rogozin, Alexander Vakhrushev, Anton Savchenko, Andrey Takáč, Martin Gasnikov, Alexander |
| contents | The objective of Vertical Federated Learning (VFL) is to collectively train a model using features available on different devices while sharing the same users. This paper focuses on the saddle point reformulation of the VFL problem via the classical Lagrangian function. We first demonstrate how this formulation can be solved using deterministic methods. More importantly, we explore various stochastic modifications to adapt to practical scenarios, such as employing compression techniques for efficient information transmission, enabling partial participation for asynchronous communication, and utilizing coordinate selection for faster local computation. We show that the saddle point reformulation plays a key role and opens up possibilities to use mentioned extension that seem to be impossible in the standard minimization formulation. Convergence estimates are provided for each algorithm, demonstrating their effectiveness in addressing the VFL problem. Additionally, alternative reformulations are investigated, and numerical experiments are conducted to validate performance and effectiveness of the proposed approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_15996 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Exploring New Frontiers in Vertical Federated Learning: the Role of Saddle Point Reformulation Beznosikov, Aleksandr Kormakov, Georgiy Grigorievskiy, Alexander Rudakov, Mikhail Nazykov, Ruslan Rogozin, Alexander Vakhrushev, Anton Savchenko, Andrey Takáč, Martin Gasnikov, Alexander Optimization and Control Machine Learning The objective of Vertical Federated Learning (VFL) is to collectively train a model using features available on different devices while sharing the same users. This paper focuses on the saddle point reformulation of the VFL problem via the classical Lagrangian function. We first demonstrate how this formulation can be solved using deterministic methods. More importantly, we explore various stochastic modifications to adapt to practical scenarios, such as employing compression techniques for efficient information transmission, enabling partial participation for asynchronous communication, and utilizing coordinate selection for faster local computation. We show that the saddle point reformulation plays a key role and opens up possibilities to use mentioned extension that seem to be impossible in the standard minimization formulation. Convergence estimates are provided for each algorithm, demonstrating their effectiveness in addressing the VFL problem. Additionally, alternative reformulations are investigated, and numerical experiments are conducted to validate performance and effectiveness of the proposed approach. |
| title | Exploring New Frontiers in Vertical Federated Learning: the Role of Saddle Point Reformulation |
| topic | Optimization and Control Machine Learning |
| url | https://arxiv.org/abs/2602.15996 |