On Reinforcement Learning for Full-Length Game of StarCraft.
Zhen-Jia PangRuo-Ze LiuZhou-Yu MengYi ZhangYang YuTong LuPublished in: AAAI (2019)
Keyphrases
- reinforcement learning
- video games
- opponent modeling
- game play
- computer games
- game playing
- game theory
- model free
- human players
- multi agent
- nash equilibrium
- game design
- function approximation
- robotic control
- educational games
- temporal difference learning
- reinforcement learning algorithms
- serious games
- monte carlo tree search
- game development
- imperfect information
- learning algorithm
- fixed length
- state space
- reinforcement learning agents
- optimal policy
- transfer learning
- real time strategy games
- machine learning
- board game
- reinforcement learning methods
- learning problems
- neural network
- ai systems
- function approximators
- game based learning
- temporal difference
- learning process
- optimal strategy
- maximum number
- dynamic programming
- supervised learning
- markov decision processes