Regularized Best Responses and Reinforcement Learning in Games.
Panayotis MertikopoulosWilliam H. SandholmPublished in: CoRR (2014)
Keyphrases
- reinforcement learning
- action sets
- learning agents
- state space
- reinforcement learning algorithms
- model free
- function approximation
- reinforcement learning agents
- least squares
- computer games
- video games
- learning algorithm
- optimal policy
- game theory
- robotic control
- game theoretic
- policy search
- nash equilibrium
- markov decision processes
- learning process
- multi agent reinforcement learning
- stochastic games
- imperfect information
- coalitional games
- total least squares
- perfect information
- game playing
- markov decision process
- game tree
- action selection
- game play
- educational games
- learning problems
- supervised learning
- cooperative games
- weighted majority
- online game
- multiagent learning
- game players
- board game
- human computation
- risk minimization
- sensory inputs
- regularized least squares
- regularization method
- learning agent
- nash equilibria
- neural network
- temporal difference
- optimal control
- cooperative
- machine learning