Closed-Loop Adaptive Reinforcement Learning

I develop policy-gradient and adaptive-learning methods that update controllers directly inside feedback loops while preserving stability, robustness, and performance guarantees for dynamical systems.

Publications

Feiran Zhao, Florian Dorfler, Alessandro Chiuso, Keyou You. "Data-Enabled Policy Optimization for Direct Adaptive Learning of the LQR." IEEE Transactions on Automatic Control (full paper), 70(11):7217-7232, 2025.

Feiran Zhao, Alessandro Chiuso, Florian Dorfler. "Policy Gradient Adaptive Control for the LQR: Indirect and Direct Approaches." IEEE Transactions on Automatic Control (full paper), Submitted, 2025.

Felix Laurent, Feiran Zhao, Jaap Eising, Florian Dorfler. "Adaptive Control of Unknown Linear Switched Systems via Policy Gradient Methods." European Control Conference (ECC), Accepted, 2026.

Feiran Zhao, Keyou You. "Policy Gradient Methods for the Cost-Constrained LQR: Strong Duality and Global Convergence." IEEE Transactions on Automatic Control, 70(12):8414-8421, 2025.

Feiran Zhao, Xingyu Sha, Keyou You. "Asynchronous Parallel Policy Gradient Methods for the Linear Quadratic Regulator." IEEE Transactions on Automatic Control, 70(7):4920-4927, 2025.

Feiran Zhao, Xingyun Fu, Keyou You. "Convergence and Sample Complexity of Policy Gradient Methods for Stabilizing Linear Systems." IEEE Transactions on Automatic Control (full paper), 70(3):1455-1466, 2025.

Feiran Zhao, Xingyun Fu, Keyou You. "Globally convergent policy gradient methods for linear quadratic control of partially observed systems." IFAC-PapersOnLine, 56(2):5506-5511, 2023.

Feiran Zhao, Keyou You, Tamer Basar. "Global Convergence of Policy Gradient Primal-Dual Methods for Risk-Constrained LQRs." IEEE Transactions on Automatic Control (full paper), 68(5):2934-2949, 2023.

Feiran Zhao, Keyou You. "Primal-dual learning for the model-free risk-constrained linear quadratic regulator." Learning for Dynamics and Control (L4DC), pp. 702-714, 2021.

Feiran Zhao, Keyou You, Tamer Basar. "Infinite-horizon Risk-constrained Linear Quadratic Regulator with Average Cost." IEEE Conference on Decision and Control (CDC), pp. 390-395, 2021.

Feiran Zhao, Keyou You. "Minimax Q-learning control for linear systems using the Wasserstein metric." Automatica, 149:110850, 2023.