Skip to content

Evolutionary Stability Does Not Guarantee Learning Accessibility: A Multi-Agent Reinforcement Learning Perspective on Cooperation Emergence

Sep 2026 · arXiv.org · Vol abs/2609.27664 · 0 citations · 35 references
Computer Science

TL;DR

The results indicate that evolutionary stability and learning accessibility are distinct properties of a coupled game--learning system in a transparent three-agent governance-motivated game involving a government, a platform firm, and users.

Abstract

Cooperation emergence is a central problem in multi-agent systems because decentralized agents must coordinate while adapting to the changing behavior of others. Evolutionary game theory identifies strategically stable outcomes, but stability under a population adjustment dynamic need not imply that finite-sample learning agents can reach the same outcome through local reward feedback. We study this distinction in a transparent three-agent governance-motivated game involving a government, a platform firm, and users. We derive replicator dynamics for the fixed stage-game incentives, evaluate the cooperative evolutionary basin on a symmetric initial-condition grid, and compare it with learning-basin estimates for three decentralized value-based learners. The learning analysis uses independent Q-learning with $\varepsilon$-greedy action selection, scaled Boltzmann exploration, and SA--EA BQL under the same payoff environment and outcome criterion. The evolutionary basin has volume $V_E=1.00$ on the sampled grid. The empirical learning basin is $0.88$ for $\varepsilon$-IQL and $0.00$ for both scaled Boltzmann and SA--EA BQL. Diagnostic traces show that broader action diversity and nonzero value separation can coexist with failure to sustain the cooperative joint action in this fixed configuration. These results indicate that evolutionary stability and learning accessibility are distinct properties of a coupled game--learning system. The shared-bike setting is a motivating application; the broader contribution is a framework for comparing population-level stability with the finite-sample accessibility of cooperation under specified multi-agent learning dynamics.

View source

Similar papers

Preprint Aug 2026

Emergence of cooperation: A reputation-modulated reinforcement learning

The results reveal that reputation-modulated learning significantly promotes the emergence of cooperative behavior, and the discontinuous phase transition from full cooperation to full defection as the temptation increases.

Chenyang Zhao, Ji-Qiang Zhang, Li Chen et al. · 0 citations
Sep 2026

Regime Structure in Adaptive Tariff Conflicts: A Multi-Agent Reinforcement Learning Analysis

Adaptive strategic conflicts unfold through repeated, history-dependent interactions that conventional equilibrium analysis captures only imperfectly. We present a multi-agent reinforcement learning (MARL) phase-diagram methodology that maps structural parameter spaces to stable outcome regimes and coordination-sensiti...

Nikhil Muthukumar, Micah Chiang, Jeffrey Chen et al. · 0 citations
Preprint Aug 2026

Evolution of cooperation with Q-learning: how much information do we need?

Mechanistic analyses show that a moderate neighborhood size enables individuals to strike an optimal balance between information sufficiency and decision-making tractability, which allows them to detect reciprocal opportunities while avoiding the deterioration of decision quality due to information overload.

Yi-Hsin Ku, Xin Ou, Ji-Qiang Zhang et al. · 0 citations
Preprint Oct 2026

After Cooperation Is Learned: Gradient Routing and Optimizer-Dependent Maintenance in Multi-Agent Reinforcement Learning

This work studies cooperation maintenance, defined as the survival of a behaviorally verified cooperative policy under continued training, and identifies a conditional, scale-sensitive maintenance risk associated with direct value-gradient routing rather than a universal failure of critics.

Chao-Yuan Hao, Wen-Tao Yue, Tian-You Lai et al. · 0 citations
Preprint Sep 2026

Entropic Risk-Sensitive Evolutionary Learning and Equilibrium Selection in Coordination Games

We study risk-sensitive evolutionary learning dynamics and their long-run equilibrium selection behaviors in coordination games. Agents'risk attitudes enter through the classical entropic risk measure, which evaluates opponent-induced payoff uncertainty and feeds into noisy best responses under two standard revision pr...

Solaleh Mohammadi, Xiang Gao, Kai-Qing Zhang · 0 citations
#artificial intelligence Preprint Sep 2026

An Evolutionary Computation Framework for Multi-Agent Q-Learning with Mean-Field Environmental Feedback

A coupled learning--environment model in which agents update stateless $Q-values on a fixed graph, while their population-average behavior drives an environmental variable that dynamically modifies the payoff matrix is formulated.

Li-Chen Wang, Shi-Jia Hua, Lin-Jie Liu · 0 citations

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.