Discrete linear-complexity reinforcement learning in continuous action spaces for Q-learning algorithms.
Peyman TavallaliGary B. Doran Jr.Lukas MandrakePublished in: CoRR (2018)
Keyphrases
- linear complexity
- action space
- reinforcement learning
- continuous action
- continuous state spaces
- learning algorithm
- continuous state
- policy search
- state space
- markov decision processes
- reinforcement learning algorithms
- state and action spaces
- real valued
- reinforcement learning methods
- learning problems
- control policies
- function approximation
- machine learning
- state action
- stochastic processes
- action selection
- temporal difference
- skill learning
- model free
- function approximators
- control problems
- optimal policy
- reinforcement learning problems
- transfer learning
- search algorithm
- control policy
- multi agent
- learning tasks
- bayesian networks
- learning agent
- partially observable markov decision processes
- belief state
- partially observable
- search space