Event-triggered reinforcement learning approach for unknown nonlinear continuous-time system.
Xiangnan ZhongZhen NiHaibo HeXin XuDongbin ZhaoPublished in: IJCNN (2014)
Keyphrases
- reinforcement learning
- state space
- optimal control
- markov decision processes
- markov chain
- model free
- temporal difference
- function approximation
- event detection
- dynamical systems
- machine learning
- multi agent
- markov processes
- news articles
- least squares
- action selection
- stochastic processes
- solving nonlinear
- initially unknown
- stream processing
- temporal difference learning
- stochastic differential equations
- iterative learning control
- continuous time bayesian networks