aiqbench analysis
This research advances decentralized machine learning by enabling agents to converge on socially optimal equilibria within dynamic, low-bandwidth networks without prior knowledge of the game structure. The proposed method utilizes semantic signal exchange and temporal data reconstruction to achieve finite-time logarithmic regret guarantees under various social welfare objectives. Such advancements are critical for scaling cooperative AI systems in distributed environments where communication constraints and network topology changes are prevalent.
Decentralized Optimal Equilibrium Learning Over Dynamic Networks
arXiv:2609.17601v1 Announce Type: cross Abstract: This paper studies decentralized learning of socially optimal equilibria in finite normal-form games over dynamic communication networks. Each agent observes only its own realized payoffs, does not know the game a priori, and can communicate only with time-varying neighbors using low-bandwidth messages. We propose networked decentralized optimal equilibrium learning dynamics in which agents generate randomized semantic content/discontent signals from local payoff comparisons and exchange time-stamped time-stacked tables rather than raw actions, payoff information or local estimates/parameters. The method combines table fusion with temporal majority reconstruction to mitigate dynamic communication while preserving fully decentralized operation. We establish finite-time logarithmic regret guarantees, with an in-phase exploration perturbation, for optimal equilibrium selection under utilitarian and proportional-fair social welfare objectives. Simulation results further show that the proposed approach can effectively select socially desirable equilibria over dynamic communication networks.
Abstract:This paper studies decentralized learning of socially optimal equilibria in finite normal-form games over dynamic communication networks. Each agent observes only its own realized payoffs, does not know the game a priori, and can communicate only with time-varying neighbors using low-bandwidth messages. We propose networked decentralized optimal equilibrium learning dynamics in which agents generate randomized semantic content/discontent signals from local payoff comparisons and exchange time-stamped time-stacked tables rather than raw actions, payoff information or local estimates/parameters. The method combines table fusion with temporal majority reconstruction to mitigate dynamic communication while preserving fully decentralized operation. We establish finite-time logarithmic regret guarantees, with an in-phase exploration perturbation, for optimal equilibrium selection under utilitarian and proportional-fair social welfare objectives. Simulation results further show that the proposed approach can effectively select socially desirable equilibria over dynamic communication networks.
Extended version of the paper: S. T. Kiremitci and M. O. Sayin, "Decentralized optimal equilibrium learning over dynamic networks", to appear in the Proceedings of the IEEE Conference on Decision and Control, 2026
Computer Science and Game Theory (cs.GT); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2609.17601 [cs.GT]
(or arXiv:2609.17601v1 [cs.GT] for this version)
Full text under the publisher's open licence · 187 words
Related coverage
News
Quantum Computing Report · Jul 25, 2026
DIGing--SGLD: Decentralized and Scalable Langevin Sampling over Time--Varying Networks
arXiv · Aug 25, 2026
CoRe-MARL: Cooperative Redistribution Under Unknown Dynamics Using Recurrent Multi-Agent Reinforcement Learning
arXiv · Sep 17, 2026
When Does Scale-Invariant Optimization Become Unstable? An Exact Schedule Law with Weight Decay
arXiv · Sep 9, 2026