Portrait of Ziping Xu
zipingxu@unc.edu

Ziping Xu 许子平

Assistant Professor · School of Data and Information Sciences

University of North Carolina at Chapel Hill

I am an Assistant Professor at UNC Chapel Hill, School of Data and Information Sciences.

From 2023–2025, I was a Postdoctoral Fellow in Statistics at Harvard University with Professor Susan Murphy. I obtained my PhD in Statistics from the University of Michigan, advised by Professor Ambuj Tewari. Before that, I obtained my B.S. in Data Science from Peking University in 2018, where I was advised by Professor Song Xi Chen.

My research explores new frontiers of data-driven decision making — reinforcement learning in particular — with applications to digital interventions such as mobile health.

News

Research interests

My research explores new frontiers of data-driven decision making — reinforcement learning in particular — with applications to digital interventions such as mobile health. A few highlights of my work:

  • Sample-efficient online algorithm design. Sample efficiency is crucial in data-scarce domains like mobile health. I develop algorithms that leverage structural information to improve sample efficiency [I, II, III].
  • Statistical inference for adaptively collected data. Adaptive experimental designs, commonly used in mobile health, introduce dependencies across time and users that challenge existing inference methods. I develop approaches that remain valid in these settings [I, II].
  • Transfer and multitask learning. Limited sample sizes in mobile health often make it necessary to transfer knowledge from existing datasets, but domain shifts — participant demographics, intervention designs, societal changes — complicate this. I design transfer learning approaches that account for such shifts [I, II, III, IV, V].
Mobile health clinical trial design. I am actively involved in designing reinforcement-learning-powered digital intervention solutions for real mobile health clinical trials. One such trial is ADAPTS-HCT, which targets medication adherence among adolescents and young adults undergoing bone marrow transplantation. I lead the RL algorithm design for adaptive delivery of digital interventions as part of the intervention package. See my talk at the mDOT webinar series for more details.