Applying supervised and reinforcement learning methods to create neural-network-based agents for playing StarCraft II.
Michal OpanowiczPublished in: CoRR (2021)
Keyphrases
- reinforcement learning methods
- learning agent
- reinforcement learning
- multi agent systems
- opponent modeling
- multi agent
- dynamical systems
- reinforcement learning algorithms
- neural network
- learning algorithm
- imperfect information
- multiple agents
- action selection
- game playing
- learning capabilities
- control problems
- single agent
- action space
- state space
- dynamic environments
- supervised learning
- control strategies
- semi supervised
- decision making
- machine learning
- model free
- learning environment
- human players