Stochastic Q-learning for Large Discrete Action Spaces.
Fares FouratiVaneet AggarwalMohamed-Slim AlouiniPublished in: CoRR (2024)
Keyphrases
- continuous state spaces
- action space
- state space
- continuous state
- reinforcement learning
- markov decision processes
- real valued
- control policies
- stochastic processes
- continuous action
- action selection
- state and action spaces
- state action
- reinforcement learning methods
- control problems
- state dependent
- optimal policy
- markov decision process
- function approximators
- heuristic search
- reinforcement learning algorithms
- partially observable markov decision processes
- stochastic approximation
- markov chain
- markov decision problems
- decision making
- learning algorithm
- finite state
- partially observable
- dynamic programming
- policy search
- reinforcement learning problems
- belief state