An Intelligent Agent Playing Generic Action Games based on Deep Reinforcement Learning with Memory Restrictions.
Lucas Antunes de AlmeidaMarcelo Resende ThieloPublished in: SBGames (2020)
Keyphrases
- reinforcement learning
- action sets
- game playing
- computer games
- temporal difference learning
- action selection
- games played
- action space
- partially observable domains
- game players
- learning agents
- extensive form games
- board game
- state action
- online game
- reward shaping
- foreign language learners
- function approximation
- state space
- intelligent agents
- video games
- reinforcement learning algorithms
- imperfect information
- multi player
- game design
- temporal difference
- reinforcement learning agents
- transition model
- game theory
- model free
- learning algorithm
- card game
- optimal policy
- markov decision processes
- main memory
- stochastic games
- human players
- serious games
- dynamic programming
- nash equilibria
- memory requirements
- multiagent learning
- general game playing
- digital games
- game theoretic
- card games
- perfect information
- multi agent
- two player games
- human actions
- nash equilibrium
- mobile robot
- playing games
- computer poker
- educational games