Actor-critic continuous state reinforcement learning for wind-turbine control robust optimization.
Borja Fernández-GaunaManuel GrañaJuan-Luis Osa-AmilibiaXabier LarruceaPublished in: Inf. Sci. (2022)
Keyphrases
- reinforcement learning
- continuous state
- actor critic
- robust optimization
- optimal control
- partially observable markov decision processes
- temporal difference
- control policy
- policy gradient
- reinforcement learning algorithms
- function approximation
- approximate dynamic programming
- control strategies
- state space
- state action
- rl algorithms
- action space
- stochastic games
- mathematical programming
- model free
- learning algorithm
- action selection
- policy iteration
- control system
- machine learning
- optimal policy
- markov decision processes
- multi agent
- finite state
- partially observable
- monte carlo
- objective function
- evolutionary algorithm
- average reward
- multi agent systems
- function approximators
- neural network
- dynamical systems