Absolutely expedient algorithms for learning Nash equilibria
Consider an \(n\)-person non-zero sum game \(\Gamma\) with finite pure strategy spaces \(S_ i\) for the players \((1\leq i\leq n)\), where for each pure strategy vector \(\bar s= (s_ 1,\dots, s_ n)\) in \(S_ 1\times S_ 2\times\cdots\times S_ n\), the payoff function of the \(i\)th player \(f_ i(\bar s)\) is a realization \(x_ i\) of a random variable with a probability distribution \(F(\cdot| \bar s,i)\) on \([0,1]\). It is assumed that all the distributions \(F(\cdot| \bar s, i)\) for all \(\bar s\) and \(i\) are unknown to the players, and player \(i\) knows only the realization \(x_ i\), \(1\leq i\leq n\). Consider now such a game with incomplete information, where the players repeat playing the game \(\Gamma\) infinitely many times. The author shows that there is a ``learning algorithm, which allows all the players to improve their starting strategies in such a way that they converge to a Nash equilibrium of the game \(\Gamma\). For this purpose a team concept associated with learning automata models is employed.
- A cooperative game of a pair of learning automata
- A Learning Model for Routing in Telephone Networks
- Absolute expediency of learning automata
- Absolutely Expedient Learning Algorithms for Stochastic Automata
- Decentralized learning in finite Markov chains
- Decentralized learning of Nash equilibria in multi-person stochastic games with incomplete information
- Dynamic noncooperative game theory
- epsilon-optimality of a general class of learning algorithms
- Games with Incomplete Information Played by “Bayesian” Players, I–III Part I. The Basic Model
- scientific article; zbMATH DE number 3878095 (Why is no real title available?)
- scientific article; zbMATH DE number 4043678 (Why is no real title available?)
- scientific article; zbMATH DE number 41338 (Why is no real title available?)
- scientific article; zbMATH DE number 47865 (Why is no real title available?)
- scientific article; zbMATH DE number 52527 (Why is no real title available?)
- scientific article; zbMATH DE number 3599745 (Why is no real title available?)
- Learning Algorithms for Two-Person Zero-Sum Stochastic Games with Incomplete Information: A Unified Approach
- Learning Optimal Discriminant Functions through a Cooperative Game of Automata
- Relaxation Labeling with Learning Automata
- Learning algorithms for repeated bimatrix Nash games with incomplete information
- Learning through reinforcement and replicator dynamics
- A strategic learning algorithm for state-based games
- Finding multiple Nash equilibria via machine learning-supported Gröbner bases
- Learning convex partitions and computing game-theoretic equilibria from best response queries
- Verified learning without regret. From algorithmic game theory to distributed systems with mechanized complexity guarantees
- Global Nash convergence of Foster and Young's regret testing
- Transient and asymptotic dynamics of reinforcement learning in games
- On learning algorithms for Nash equilibria
- An N-player sequential stochastic game with identical payoffs
- Learning behavior of variable-structure stochastic automata in a three-person zero-sum game
- Distributed learning of the global maximum in a two-player stochastic game with identical payoffs
- A note on bimatrix games with an unknown payoff matrix
- Learning automata in feedforward connectionist systems
- Specification-Guided Learning of Nash Equilibria with High Social Welfare
This page was built for publication: Absolutely expedient algorithms for learning Nash equilibria
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1322380)
