Fanyu Que

About

Fanyu Que is a leading researcher in safe reinforcement learning, with a primary focus on developing algorithms that balance reward maximization with critical safety constraints. Their most influential work, "Accelerated Primal-Dual Policy Optimization for Safe Reinforcement Learning" (2018), has garnered 65 citations and addresses the fundamental challenge of training agents to operate reliably in high-stakes environments. Que's research centers on the Constrained Markov Decision Process (CMDP) framework, where they pioneered accelerated primal-dual optimization methods that significantly improve both convergence speed and constraint satisfaction. This work has direct implications for real-world applications such as autonomous driving, robotics, and healthcare systems, where unsafe actions can have severe consequences. By introducing more efficient algorithms for solving CMDPs, Que has helped bridge the gap between theoretical reinforcement learning and practical deployment in safety-critical domains. Their contributions continue to influence a growing body of research on trustworthy AI systems, making Que a notable figure in the advancement of responsible and robust machine learning.

Research Focus

Key Achievements

Top Papers

1. 1
   Accelerated Primal-Dual Policy Optimization for Safe Reinforcement Learning

   65 citations · 2018

Key Collaborators

Contact & Links