Tairan He

Papers

Total Citations

H-Index

About

Tairan He is a researcher specializing in safe reinforcement learning, control theory, and autonomous robotics — fields at the critical intersection of artificial intelligence and real-world deployment. His work addresses one of the most pressing challenges in modern AI: ensuring that learning-based systems behave safely and reliably when transferred from simulation to physical environments. He has made notable contributions to the theoretical foundations of safe RL, most prominently through his highly cited survey "State-wise Safe Reinforcement Learning" (2023, 41 citations), which has become a key reference for researchers navigating the landscape of constraint-satisfaction methods in RL. His parallel work on safety index synthesis, leveraging sum-of-squares programming (2023, 21 citations), offers a rigorous mathematical framework for designing safety-aware control policies under real-world constraints — a technically demanding problem that previously required extensive expert intuition. More recently, He has extended this work toward adaptive autonomous systems, exploring how robots can rapidly adjust to dynamic, uncertain environments without sacrificing safety guarantees, as demonstrated in "Safe Deep Policy Adaptation" (2024). Together, these contributions position Tairan He as an emerging voice in trustworthy autonomy, bridging formal control theory with modern deep learning to make intelligent systems both capable and dependably safe.