Convergence of Reinforcement Learning to Nash Equilibrium: a search market - Archive ouverte HAL Access content directly
Journal Articles Physica A: Statistical Mechanics and its Applications Year : 2005

Convergence of Reinforcement Learning to Nash Equilibrium: a search market

(1) , (2, 3)
1
2
3

Abstract

Since the introduction of Reinforcement Learning (RL) in Game Theory, a growing literature is concerned with the theoretical convergence of RL-driven outcomes towards Nash equilibrium. In this paper, we apply this issue to a search-theoretic framework (posted-price market) where sellers are confronted with a population of imperfectly informed buyers and take one decision per period (posted prices) with no direct interactions between sellers. We focus on three different scenarios with varying buyers' characteristics. For each of these scenarios, we quantitatively and qualitatively test whether the learned variable (price strategy) converges to the Nash equilibrium. We also study the impact of the temperature parameter (defining the exploitation/exploration trade off) on these results.
Fichier principal
Vignette du fichier
Darmon2005.pdf (164.71 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02167937 , version 1 (29-04-2022)

Licence

Attribution - NonCommercial - CC BY 4.0

Identifiers

Cite

Eric Darmon, Roger Waldeck. Convergence of Reinforcement Learning to Nash Equilibrium: a search market. Physica A: Statistical Mechanics and its Applications, 2005, 355 (1), pp.119-130. ⟨10.1016/j.physa.2005.02.074⟩. ⟨hal-02167937⟩
84 View
20 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More