Modulation-Enhanced Excitation for Continuous-Time Reinforcement Learning via Symmetric Kronecker Products.
Brent A. WallaceJennie SiPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- state space
- optimal control
- function approximation
- markov chain
- machine learning
- model free
- markov decision processes
- kronecker product
- neural network
- reinforcement learning algorithms
- optimal policy
- dynamical systems
- reinforcement learning methods
- temporal difference
- dynamic programming
- markov processes
- robotic control
- signal to noise ratio
- power system
- action selection
- production line
- learning algorithm