Welcome to Chen Yan's Page

Research Interest: 

• I develop theory and algorithms for large-scale stochastic optimization and control. On the first-order front, I proved the exponential-rate asymptotic optimality of the Whittle index policy and established necessary and sufficient conditions for achieving this rate. I also developed a second-order, variance-aware method that is tractable and provably improves upon traditional fluid/mean-field approaches. 

• My current research focuses on extending these ideas to heterogeneous systems, efficient algorithm design and implementation, and learning under model uncertainty, with applications to wireless scheduling, matching platforms, ride-hailing, and healthcare operations.

Selected Publications: 

• Chen Yan, Weina Wang, Lei Ying, "Achieving Õ(1/N) Optimality Gap in Restless Bandits through Gaussian Approximation" ,  (NeurIPS 2025 Spotlight) [arXiv] 

• Chen Yan, "An Optimal-Control Approach to Infinite-Horizon Restless Bandits Achieving Asymptotic Optimality with Minimal Assumptions" , (Conference on Decision and Control,  Dec 2024) [arXiv]

• Chen Yan, Nicolas Gast and Bruno Gaujal, "LP-Based Policies for Restless Bandits: Necessary and Sufficient Conditions for (Exponentially Fast) Asymptotic Optimality",  (Mathematics of Operations Research, Dec 2023) [arXiv]

• Chen Yan, Nicolas Gast and Bruno Gaujal, "Exponential Asymptotic Optimality of Whittle Index Policy",  (Queueing Systems, May 2023)  [arXiv] 

Preprints: 

• Chen Yan, Nicolas Gast and Bruno Gaujal, "Reoptimization Nearly Solves Weakly Coupled MDPs",  [arXiv] 

• Stéphane Girard, Thomas Opitz, Antoine Usseglio-Carleve, Chen Yan, "Changepoint Identification in Heavy-Tailed Distributions",  [HAL] 

• Chen Yan, Alexandre REIFFERS-MASSON, "Certainty Equivalence Control-Based Heuristics in Multi-Stage Convex Stochastic Optimization Problems",  [arXiv]