Personalized and Resilient Distributed Learning Through Opinion Dynamics
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915700399931392 |
|---|---|
| author | Ballotta, Luca Bastianello, Nicola Ferrari, Riccardo M. G. Johansson, Karl H. |
| author_facet | Ballotta, Luca Bastianello, Nicola Ferrari, Riccardo M. G. Johansson, Karl H. |
| contents | In this paper, we address two practical challenges of distributed learning in multi-agent network systems, namely personalization and resilience. Personalization is the need of heterogeneous agents to learn local models tailored to their own data and tasks, while still generalizing well; on the other hand, the learning process must be resilient to cyberattacks or anomalous training data to avoid disruption. Motivated by a conceptual affinity between these two requirements, we devise a distributed learning algorithm that combines distributed gradient descent and the Friedkin-Johnsen model of opinion dynamics to fulfill both of them. We quantify its convergence speed and the neighborhood that contains the final learned models, which can be easily controlled by tuning the algorithm parameters to enforce a more personalized/resilient behavior. We numerically showcase the effectiveness of our algorithm on synthetic and real-world distributed learning tasks, where it achieves high global accuracy both for personalized models and with malicious agents compared to standard strategies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_14081 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Personalized and Resilient Distributed Learning Through Opinion Dynamics Ballotta, Luca Bastianello, Nicola Ferrari, Riccardo M. G. Johansson, Karl H. Multiagent Systems Machine Learning Signal Processing Optimization and Control In this paper, we address two practical challenges of distributed learning in multi-agent network systems, namely personalization and resilience. Personalization is the need of heterogeneous agents to learn local models tailored to their own data and tasks, while still generalizing well; on the other hand, the learning process must be resilient to cyberattacks or anomalous training data to avoid disruption. Motivated by a conceptual affinity between these two requirements, we devise a distributed learning algorithm that combines distributed gradient descent and the Friedkin-Johnsen model of opinion dynamics to fulfill both of them. We quantify its convergence speed and the neighborhood that contains the final learned models, which can be easily controlled by tuning the algorithm parameters to enforce a more personalized/resilient behavior. We numerically showcase the effectiveness of our algorithm on synthetic and real-world distributed learning tasks, where it achieves high global accuracy both for personalized models and with malicious agents compared to standard strategies. |
| title | Personalized and Resilient Distributed Learning Through Opinion Dynamics |
| topic | Multiagent Systems Machine Learning Signal Processing Optimization and Control |
| url | https://arxiv.org/abs/2505.14081 |