Adaptive Policy Regularization for Offline to Online Reinforcement Learning in HVAC Control

Watch an 18-minute conference talk from ACM exploring how adaptive policy regularization can be applied to offline-to-online reinforcement learning specifically for HVAC control systems. Learn about the research conducted by authors Hsin-Yu Liu, Bharathan Balaji, Rajesh Gupta, and Dezhi Hong as they present their findings on improving HVAC control mechanisms through machine learning techniques. Discover the methodology behind transitioning from offline to online reinforcement learning while maintaining system stability and performance in building automation applications.