Quantum-enhanced reinforcement learning for finite-episode games with discrete state spaces.
Florian NeukartDavid Von DollenChristian SeidelGabriele CompostellaPublished in: CoRR (2017)
Keyphrases
- state space
- reinforcement learning
- continuous state spaces
- reinforcement learning algorithms
- state and action spaces
- action sets
- finite number
- markov decision processes
- continuous state
- action space
- optimal policy
- learning agents
- state abstraction
- reinforcement learning agents
- markov chain
- heuristic search
- function approximation
- continuous state and action spaces
- control problems
- temporal difference
- dynamic programming
- markov decision process
- continuous domains
- planning problems
- reinforcement learning methods
- game theory
- video games
- stochastic games
- quantum computing
- real numbers
- computer games
- digital games
- real valued functions
- policy search
- arbitrary size
- ridgelet transform
- quantum computation
- partially observable
- model free
- game play
- learning algorithm
- temporal difference learning
- imperfect information
- continuous functions
- multiagent learning
- learning agent
- event sequences
- initial state
- machine learning
- game theoretic
- game design
- educational games
- pattern discovery