Sample Path Sharing in Policy Improvement for Indoor Air Temperature Control.
Di WuQing-Shan JiaPublished in: WODES (2014)
Keyphrases
- sample path
- temperature control
- policy iteration
- asymptotic analysis
- average reward
- fluid model
- optimal policy
- control algorithm
- markov chain
- markov decision processes
- lost sales
- large deviations
- least squares
- markov decision process
- steady state
- state dependent
- reinforcement learning
- asymptotically optimal
- long run
- model free
- finite state
- real time
- pid controller
- control method
- temporal difference
- infinite horizon
- fixed point
- dynamical systems
- dynamic programming
- objective function
- genetic algorithm