MEMOA: Massive Mixtures of Online Agents via Mean-Field Decentralized Nash Equilibria
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918486936125440 |
|---|---|
| author | Yang, Xuwei Emerson, David B. Tavakoli, Fatemeh Kratsios, Anastasis |
| author_facet | Yang, Xuwei Emerson, David B. Tavakoli, Fatemeh Kratsios, Anastasis |
| contents | In the modern age of large-scale AI, federated learning has become an increasingly important tool for training large populations of AI agents; however, its computational and communication costs can rapidly fail to scale with the number of agents. This is precisely where decentralized agentic strategies shine: each agent acts autonomously, using only its own state together with a minimal summary of the ensemble, namely the mean-field. We derive the unique optimal decentralized policy in closed form. Optimality is characterized through a worst-client/minimax criterion: minimizing the under-performer regret, namely the maximal online cost incurred by the weakest agent in the ensemble. We further prove that the resulting decentralized policy asymptotically converges, in the large-population limit, to the Nash-optimal centralized policy, whose direct computation is not scalable. We use an online weighting mechanism to optimize the server-computed mixture of client predictions, thereby improving the mean prediction in addition to the previously optimized weakest-client prediction. Numerical experiments verify our theoretical guarantees and demonstrate that our decentralized policy typically outperforms natural greedy decentralized baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_05492 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MEMOA: Massive Mixtures of Online Agents via Mean-Field Decentralized Nash Equilibria Yang, Xuwei Emerson, David B. Tavakoli, Fatemeh Kratsios, Anastasis Machine Learning 91A80, 91A16, 93E20, 49N10 I.2.11; I.2.8 In the modern age of large-scale AI, federated learning has become an increasingly important tool for training large populations of AI agents; however, its computational and communication costs can rapidly fail to scale with the number of agents. This is precisely where decentralized agentic strategies shine: each agent acts autonomously, using only its own state together with a minimal summary of the ensemble, namely the mean-field. We derive the unique optimal decentralized policy in closed form. Optimality is characterized through a worst-client/minimax criterion: minimizing the under-performer regret, namely the maximal online cost incurred by the weakest agent in the ensemble. We further prove that the resulting decentralized policy asymptotically converges, in the large-population limit, to the Nash-optimal centralized policy, whose direct computation is not scalable. We use an online weighting mechanism to optimize the server-computed mixture of client predictions, thereby improving the mean prediction in addition to the previously optimized weakest-client prediction. Numerical experiments verify our theoretical guarantees and demonstrate that our decentralized policy typically outperforms natural greedy decentralized baselines. |
| title | MEMOA: Massive Mixtures of Online Agents via Mean-Field Decentralized Nash Equilibria |
| topic | Machine Learning 91A80, 91A16, 93E20, 49N10 I.2.11; I.2.8 |
| url | https://arxiv.org/abs/2605.05492 |