Near-continuous time Reinforcement Learning for continuous state-action spaces.
Lorenzo CroissantMarc AbeilleBruno BouchardPublished in: ALT (2024)
Keyphrases
- continuous state
- action space
- state space
- reinforcement learning
- stochastic processes
- markov decision processes
- markov chain
- policy search
- control policies
- optimal control
- continuous state spaces
- real valued
- finite state
- function approximators
- function approximation
- optimal policy
- robot navigation
- dynamical systems
- dynamic programming
- heuristic search
- reinforcement learning algorithms
- state action
- continuous action
- state variables
- learning algorithm
- single agent
- belief state
- partially observable
- action selection
- initial state
- control strategies
- model free
- planning problems
- markov decision problems
- model checking
- multi agent systems
- multi agent