Multi-player H∞ Differential Game using On-Policy and Off-Policy Reinforcement Learning.
Peiliang AnMushuang LiuYan WanFrank L. LewisPublished in: ICCA (2020)
Keyphrases
- multi player
- reinforcement learning
- game playing
- optimal policy
- online game
- game play
- policy iteration
- markov decision process
- educational games
- action selection
- multi agent
- stochastic games
- average reward
- markov decision processes
- function approximation
- reward function
- infinite horizon
- model free
- dynamic programming
- video games
- state space
- computer games
- game tree
- subgame perfect equilibrium
- optimal control
- learning algorithm
- imperfect information
- reinforcement learning algorithms
- temporal difference
- serious games
- role playing game
- game design
- partially observable markov decision processes
- learning experience
- game based learning
- solution concepts
- long run