Larry Yan

About

Larry Yan is a rising researcher at the forefront of reinforcement learning (RL) and its application to real-world control systems, particularly in building energy optimization. His work addresses a critical challenge in the field: the brittleness of RL agents, which often overfit to training environments and fail to generalize. In his highly cited 2024 paper, "Active Reinforcement Learning for Robust Building Control," Yan introduces an active learning framework that enables RL agents to strategically query for informative experiences, dramatically improving their robustness and transferability across diverse building conditions. While still early in his career, his research has already garnered attention for tackling the practical deployment gap between RL's success in simulated games and its reliability in physical infrastructure. By bridging the gap between theoretical RL and robust, real-world control, Yan is paving the way for more adaptive and energy-efficient smart buildings. His work represents a significant step toward making RL a trustworthy tool for critical infrastructure, and he is quickly establishing himself as a key voice in the intersection of machine learning and sustainable engineering.

Research Focus

Key Achievements

Top Papers

1. 1
   Active Reinforcement Learning for Robust Building Control

   3 citations · 2024

Key Collaborators

Contact & Links