Learning to Shoot in First Person Shooter Games by Stabilizing Actions and Clustering Rewards for Reinforcement Learning.
Frank G. GlavinMichael G. MaddenPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- learning algorithm
- unsupervised learning
- action selection
- learning process
- learning problems
- action sets
- learning agents
- markov decision processes
- partially observable
- function approximation
- clustering algorithm
- reward function
- multiagent learning
- temporal difference learning
- learned knowledge
- video games
- supervised learning
- reinforcement learning algorithms
- perceptual aliasing
- state space
- autonomous learning
- learning agent
- computer games
- state action
- optimal policy
- clustering method
- model free
- partial observability
- learning capabilities
- state abstraction
- policy search
- multiagent reinforcement learning
- macro actions
- learning environment
- machine learning