Publications

• Online switching control with stability and regret guarantees
  Yingying Li, James A. Preiss, Na Li, Yiheng Lin, Adam Wierman, Jeff Shamma
  5th Annual Learning for Dynamics & Control Conference (L4DC), 2023. Oral [PDF]

• Non-asymptotic System Identification for Linear Systems with Nonlinear Policies
  Yingying Li, Tianpeng Zhang, Subhro Das, Jeff Shamma, Na Li
  The 22nd World Congress of the International Federation of Automatic Control (IFAC WC), 2023.[PDF]

• Online Hyperparameter Optimization for Class-Incremental Learning
  Yaoyao Liu, Yingying Li, Bernt Schiele, Qianru Sun
  Thirty-seventh AAAI Conference on Artificial Intelligence (AAAI), 2023. Oral [PDF]

• Safe Adaptive Learning for Linear Quadratic Regulators with Constraints
  Yingying Li, Tianpeng Zhang, Subhro Das, Jeff Shamma, Na Li
  Under review. [PDF]

• Distributed Reinforcement Learning for Decentralized Linear Quadratic Control: A Derivative-Free Policy Optimization Approach
  Yingying Li, Yujie Tang, Runyu Zhang, Na Li (the first two authors contribute equally)
  To appear in IEEE Transactions on Automatic Control (TAC), 2022. [PDF]

• Online Optimal Control with Affine Constraints
  Yingying Li, Subhro Das, Na Li
  Thirty-fifth AAAI Conference on Artificial Intelligence (AAAI), 2021. [arXiv]

• Online Learning and Distributed Control for Residential Demand Response
  Xin Chen, Yingying Li, Jun Shimada, Na Li
  IEEE Transactions on Smart Grid (TSG), 2021. [PDF]

• On the Regret Analysis of Online LQR Control with Predictions
  Runyu Zhang, Yingying Li, Na Li
  The American Control Conference (ACC), 2021. [PDF]

• Online Optimization with Predictions and Switching Costs: Fast Algorithms and the Fundamental Limit
  Yingying Li, Guannan Qu, Na Li
  IEEE Transactions on Automatic Control (TAC), 2021. [PDF]

• Leveraging Predictions in Smoothed Online Convex Optimization via Gradient-based Algorithms
  Yingying Li, Na Li
  Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS), 2020. [PDF]

• A Reliability-aware Multi-armed Bandit Approach to Learn and Select Users in Demand Response
  Yingying Li, Qinran Hu, Na Li
  Automatica Volume 119, 2020 (Editor’s Choice) [PDF] [arXiv]

• Distributed Reinforcement Learning for Decentralized Linear Quadratic Control: A Derivative-Free Policy Optimization Approach
  Yingying Li, Yujie Tang, Runyu Zhang, Na Li (the first two authors contribute equally)
  2nd Annual Conference on Learning for Dynamics and Control

• Online Optimal Control with Linear Dynamics and Predictions: Algorithms and Regret Analysis
  Yingying Li, Xin Chen, Na Li
  Thirty-third Conference on Neural Information Processing Systems (NeurIPS), 2019.
  [PDF] [Code] [Slides]

• Online Markov Decision Processes with Time-varying Transition Probabilities and Rewards
  Yingying Li, Aoxiao Zhong, Guannan Qu, Na Li
  Real-world Sequential Decision Making workshop at ICML 2019. [PDF]

• Online Learning for Markov Decision Processes in Nonstationary Environments: A Dynamic Regret Analysis
  Yingying Li, Na Li
  The American Control Conference (ACC), 2019. [PDF]

• Learning and Selecting the Right Customers for Reliability: A Multi-armed Bandit Approach
  Yingying Li, Qinran Hu, Na Li
  57th IEEE Conference on Decision and Control (CDC), 2018. [PDF]

• Using Predictions in Online Optimization with Switching Costs: A Fast Algorithm and A Fundamental Limit
  Yingying Li, Guannan Qu, Na Li
  The American Control Conference (ACC), 2018. [PDF]

• Mechanism Design for Reliability in Demand Response with Uncertainty
  Yingying Li, Na Li
  The American Control Conference (ACC), 2017. [PDF]

• Growing Optimal Scale-free Networks via Likelihood
  Michael Small, Yingying Li, Thomas Stemler, Kevin Judd
  Physical Review E 91, no. 4 (2015): 042801 [PDF]