Regularized Gradient Descent Ascent for Two-Player Zero-Sum Markov Games.
Sihan ZengThinh T. DoanJustin RombergPublished in: CoRR (2022)
Keyphrases
- markov games
- markov decision processes
- multiagent reinforcement learning
- reinforcement learning algorithms
- markov decision process
- reinforcement learning
- control problems
- objective function
- multiagent systems
- cost function
- least squares
- finite state
- loss function
- stochastic games
- state space
- multi agent
- model free
- nash equilibrium
- cooperative
- optimal policy
- function approximation
- policy iteration
- infinite horizon
- learning algorithm
- search space
- lower bound
- optimal stopping