An adjusted payoff-based procedure for normal form games

DOI10.1287/MOOR.2016.0785zbMATH Open1349.91062arXiv1106.5596OpenAlexW1625953579MaRDI QIDQ2833112FDOQ2833112

Authors: Mario Bravo 

Publication date: 16 November 2016

Published in: Mathematics of Operations Research (Search for Journal in Brave)

Abstract: We study a simple adaptive model in the framework of an N -player normal form game. The model consists of a repeated game where the players only know their own action space and their own payoff scored at each stage, not those of the other agents. Each player, in order to update her mixed action, computes the average vector payoff she has obtained by using the number of times she has played each pure action. The resulting stochastic process is analyzed via the ODE method from stochastic approximation theory. We are interested in the convergence of the process to rest points of the related continuous dynamics. Results concerning almost sure convergence and convergence with positive probability are obtained and applied to a traffic game. We also provide some examples where convergence occurs with probability zero.

Full work available at URL: https://arxiv.org/abs/1106.5596

Recommendations

• ON THE PAYOFF REPRESENTATIONS OF NORMAL FORM GAMES
• Refinements of rationalizability for normal-form games
• The theory of normal form games from the differentiable viewpoint
• The refined best-response correspondence in normal form games
• On the core of normal form games with a continuum of players
• Essential equilibrium in normal-form games with perturbed actions and payoffs
• Decompositions and potentials for normal form games
• Quantal response equilibria for normal form games
• Essential equilibria in normal-form games
• Normal form structures in extensive form games

zbMATH Keywords

adaptive dynamicslearningstochastic approximationnormal form games

Mathematics Subject Classification ID

Stochastic approximation (62L20) Noncooperative games (91A10) (n)-person games, (n>2) (91A06) Multistage and repeated games (91A20) Stochastic learning and adaptive control (93E35) Rationality and learning in game theory (91A26)

Cites Work

• A Stochastic Approximation Method
• Title not available (Why is that?)
• Potential games
• A Simple Adaptive Procedure Leading to Correlated Equilibrium
• The statistical mechanics of strategic interaction
• Adaptive game playing using multiplicative weights
• Nonconvergence to unstable points in urn models and stochastic approximations
• On the convergence of reinforcement learning
• The logit-response dynamics
• Revisiting log-linear learning: asynchrony, completeness and payoff-based implementation
• A payoff-based learning procedure and its application to traffic games
• Title not available (Why is that?)
• Learning through reinforcement and replicator dynamics
• Urn models, replicator processes, and random genetic drift
• Title not available (Why is that?)
• Learning in perturbed asymmetric games
• Individual Q-Learning in Normal Form Games
• A behavioral learning process in games
• Do stochastic algorithms avoid traps?

Cited In (5)

• Identifying behaviorally robust strategies for normal form games under varying forms of uncertainty
• Penalty-regulated dynamics and robust learning procedures in games
• Payoff-based dynamics for multiplayer weakly acyclic games
• On the stability of an adaptive learning dynamics in traffic games
• Learning and equilibrium transitions: stochastic stability in discounted stochastic fictitious play

This page was built for publication: An adjusted payoff-based procedure for normal form games

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2833112)