Compositional Reinforcement Learning for Discrete-Time Stochastic Control Systems.
Abolfazl LavaeiMateo PerezMilad KazemiFabio SomenziSadegh SoudjaniAshutosh TrivediMajid ZamaniPublished in: CoRR (2022)
Keyphrases
- reinforcement learning
- optimal control problems
- optimal control
- control system
- direct policy search
- control policies
- markov processes
- stochastic approximation
- learning automata
- control strategy
- monte carlo
- control law
- markov chain
- control problems
- function approximation
- markov decision processes
- state space
- dynamic programming
- stochastic process
- continuous state
- closed loop
- multi agent
- learning algorithm
- temporal difference
- reinforcement learning algorithms
- stochastic optimization
- robotic control
- finite state
- temporal difference learning
- continuous state spaces
- model free
- control scheme
- learning problems
- supervised learning
- markov decision process
- partially observable
- real environment
- control policy
- control strategies
- real time
- control algorithm
- machine learning
- neural network