Accelerate online reinforcement learning for building HVAC control with heterogeneous expert guidances.
Shichao XuYangyang FuYixuan WangZhuoran YangZheng O'NeillZhaoran WangQi ZhuPublished in: BuildSys@SenSys (2022)
Keyphrases
- reinforcement learning
- control problems
- optimal control
- robot control
- online learning
- control system
- multi agent
- real time
- control policy
- control strategies
- robotic control
- control method
- function approximation
- optimal policy
- control strategy
- fault diagnosis
- adaptive control
- action selection
- transfer learning
- autonomous robots
- domain experts
- temporal difference
- supervised learning
- dynamic programming
- social media