Robustness properties in fictitious-play-type algorithms
From MaRDI portal
(Redirected from Publication:4588840)
Abstract: Fictitious play (FP) is a canonical game-theoretic learning algorithm which has been deployed extensively in decentralized control scenarios. However standard treatments of FP, and of many other game-theoretic models, assume rather idealistic conditions which rarely hold in realistic control scenarios. This paper considers a broad class of best response learning algorithms, that we refer to as FP-type algorithms. In such an algorithm, given some (possibly limited) information about the history of actions, each individual forecasts the future play and chooses a (myopic) best action given their forecast. We provide a unifed analysis of the behavior of FP-type algorithms under an important class of perturbations, thus demonstrating robustness to deviations from the idealistic operating conditions that have been previously assumed. This robustness result is then used to derive convergence results for two control-relevant relaxations of standard game-theoretic applications: distributed (network-based) implementation without full observability and asynchronous deployment (including in continuous time). In each case the results follow as a direct consequence of the main robustness result.
Recommendations
Cites work
- A Class of Self-Interacting Processes with Applications to Games and Reinforced Random Walks
- A Fictitious Play Approach to Large-Scale Optimization
- A weakened form of fictitious play in two-person zero-sum games
- An iterative method of solving a game
- Asynchronous stochastic approximation with differential inclusions
- Consensus seeking in multiagent systems under dynamically changing interaction topologies
- Convergence speed in distributed consensus and averaging
- Distributed Consensus Algorithms in Sensor Networks With Imperfect Communication: Link Failures and Channel Noise
- Distributed strategic learning for wireless engineers
- Dynamic fictitious play, dynamic gradient play, and distributed convergence to Nash equilibria
- Empirical Centroid Fictitious Play: An Approach for Distributed Learning in Multi-Agent Games
- Fictitious play in \(2\times n\) games
- Fictitious play in coordination games
- Fictitious play property for games with identical interests
- Generalised weakened fictitious play
- scientific article; zbMATH DE number 1233801 (Why is no real title available?)
- scientific article; zbMATH DE number 3205074 (Why is no real title available?)
- scientific article; zbMATH DE number 3069635 (Why is no real title available?)
- Joint Strategy Fictitious Play With Inertia for Potential Games
- Learning by forgetful players
- Learning in games with strategic complementarities revisited
- Learning mixed equilibria
- Modified fictitious play for solving matrix games and linear-programming problems
- Potential games
- Probability with Martingales
- Reinforcement learning with restrictions on the action set
- Robustness properties in fictitious-play-type algorithms
- Stochastic approximation with two time scales
- Stochastic Approximations and Differential Inclusions
- Three problems in learning mixed-strategy Nash equilibria
Cited in
(3)
This page was built for publication: Robustness properties in fictitious-play-type algorithms
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4588840)
