Macro Action Selection with Deep Reinforcement Learning in StarCraft.
Sijia XuHongyu KuangZhuang ZhiRenjie HuYang LiuHuyang SunPublished in: AIIDE (2019)
Keyphrases
- action selection
- reinforcement learning
- temporal difference
- basal ganglia
- robot soccer
- continuous state and action spaces
- function approximation
- decision making
- action space
- state space
- supervised learning
- markov decision processes
- human robot
- model free
- action selection mechanism
- neural network
- reinforcement learning algorithms
- learning algorithm
- optimal control
- continuous state
- partially observable
- early stage
- transfer learning
- optimal policy