Beyond Predicting Responses: Conformal Inference for Latent Distributional Parameters
Minxing Zheng, Wenbin Zhou, and Shixiang Zhu. (2026). "Beyond Predicting Responses: Conformal Inference for Latent Distributional Parameters." arXiv:2608.03607.
CV
Education
- Ph.D. student, Statistics & Data Science — Carnegie Mellon University, 2025–present
Advisor: Prof. Shixiang Zhu - M.S., Statistics — University of Southern California, 2022–2024
Advisor: Prof. Paromita Dubey - M.S., Statistics — University of Wisconsin–Madison, 2020–2022
Advisors: Prof. Kris Sankaran and Prof. Keith Levin - B.S., Statistics; B.S., Applied Mathematics — Renmin University of China, 2017–2021
Research interests
Conformal prediction and inference; uncertainty quantification; reliable and secure machine learning; scientific AI; human–AI collaboration and decision-making.
Publications
Learning to Test: Physics-Informed Representation for Dynamical Instability Detection
Minxing Zheng, Zewei Deng, Liyan Xie, and Shixiang Zhu. (2026). "Learning to Test: Physics-Informed Representation for Dynamical Instability Detection." arXiv:2604.10967.
MISLEADER: Defending against Model Extraction with Ensembles of Distilled Models
Xueqi Cheng, Minxing Zheng, Shixiang Zhu, and Yushun Dong. (2025). "MISLEADER: Defending against Model Extraction with Ensembles of Distilled Models." arXiv:2506.02362.
Generative Conformal Prediction with Vectorized Non-Conformity Scores
Minxing Zheng and Shixiang Zhu. (2024). "Generative Conformal Prediction with Vectorized Non-Conformity Scores." arXiv:2410.13735.
Change Point Inference for Non-Euclidean Data Sequences using Distance Profiles
Paromita Dubey and Minxing Zheng. (2023; revised 2026). "Change Point Inference for Non-Euclidean Data Sequences using Distance Profiles." arXiv:2311.16025.
Interactive Visualization and Representation Analysis Applied to Glacier Segmentation
Minxing Zheng, Xinran Miao, and Kris Sankaran. (2022). "Interactive Visualization and Representation Analysis Applied to Glacier Segmentation." ISPRS International Journal of Geo-Information, 11(8), 415.
Research projects
- Deciding When to Decide: Decision-Focused Testing under Distributional Shift — working paper, 2025–present
A decision-focused testing framework for assessing whether a decision optimized in one domain remains approximately optimal after a context or feature distribution shift. - Learning to Test: Physics-Informed Representation for Dynamical Instability Detection — 2025–present
Physics-informed representation learning for efficient reliability testing under exogenous distribution shifts. - Generative Conformal Prediction with Vectorized Non-Conformity Scores — 2024–present
Adaptive conformal uncertainty sets for multimodal targets with finite-sample validity. - Derandomized Multiple Change Point Detection with FDR Control using Distance Profiles — 2025–present
Nonparametric change-point inference for metric-space objects, including extensions to multiple changes and false-discovery-rate control. - Asymptotic Behavior of the Maximum Degree Distribution under Graphon Models — 2021–2022
Bootstrap inference for maximum-degree distributions under latent graphon models.
Service & honors
- Reviewer: AISTATS; ICML
- University of Southern California Marshall School Fellowship, 2022–2024
Selected methods & applications
Distribution-free inference, generative modeling, hypothesis testing under distribution shift, model security, representation analysis, geospatial machine learning, and statistical network analysis.
Professional links
Google Scholar · GitHub · arXiv · Email