Asynchronous neurocomputing for optimal control and reinforcement learning with large state spaces.
Bruno ScherrerPublished in: Neurocomputing (2005)
Keyphrases
- optimal control
- reinforcement learning
- state space
- reinforcement learning algorithms
- dynamic programming
- control problems
- optimal policy
- markov decision processes
- feedback control
- infinite horizon
- function approximation
- class of nonlinear systems
- risk sensitive
- markov chain
- continuous state spaces
- control law
- state abstraction
- model free
- brownian motion
- actor critic
- learning algorithm
- markov decision process
- reinforcement learning methods
- optimal control problems
- partially observable
- action space
- rl algorithms
- reward function
- action selection
- control strategy
- planning problems
- dynamical systems
- policy iteration
- average cost
- linear quadratic
- continuous stirred tank reactor