
The Department of Statistics is celebrating a series of grant awards to faculty members, reflecting the breadth and impact of research taking place across the department. Supported by both the National Institutes of Health (NIH) and the National Science Foundation (NSF), these projects span topics ranging from mental health resilience and biomedical data science to generative models, simulation-based inference, and causal inference. Together, they highlight the department’s leadership in advancing statistical theory, methodology, and applications that address pressing challenges in science and society.

Dave Zhao
NIH Grant: CRCNS: Multimodal network interactions for internal state dynamics of resiliency
Professor Dave Zhao has received a National Institutes of Health (NIH) grant to investigate the biological mechanisms that allow individuals to remain resilient in the face of stress. While the harmful effects of prolonged stress are well-documented, far less is understood about why some people maintain healthy behavioral function despite adversity. This project will integrate behavioral studies, neuroimaging, electrophysiology, gene expression, and biomarker data from mouse models with advanced computational methods such as machine learning and statistical inference. By mapping how molecular, genomic, and neural circuit interactions give rise to resilience, Zhao’s work aims to establish the first comprehensive framework for understanding how resilience emerges across interconnected biological systems, with potential implications for mental health interventions.

Shulei Wang
NSF Grant: Statistical Frameworks for Self-Supervised Representation Learning and Their Biomedical Application
Professor Shulei Wang has received a National Science Foundation (NSF) grant to advance the theory and application of self-supervised representation learning—an emerging approach that leverages large amounts of unlabeled data to train machine learning models. While this method has powered breakthroughs in computer vision and natural language processing, its theoretical foundations remain poorly understood, and existing techniques cannot be directly applied to biomedical research. Wang’s project will build new mathematical frameworks to explain how self-supervised learning works and develop novel computational tools tailored to biomedical datasets such as microbiome and omics-based longitudinal data. Alongside its research contributions, the project includes educational components to engage students and the public in this growing field. By enabling more effective use of unlabeled biomedical data, these advancements could open new pathways for scientific discovery and a deeper understanding of human health.

Jingbo Liu
NSF Grant: Toward Statistically Optimal Diffusion Generative Models: Accuracy Complexity and Privacy
Professor Jingbo Liu has been awarded a National Science Foundation (NSF) grant to study the statistical foundations of diffusion models, a powerful class of generative modeling techniques that have rapidly transformed fields such as image and video synthesis, scientific simulation, and reinforcement learning. While diffusion models have shown remarkable success in practice, their theoretical limits and trade-offs remain poorly understood. Liu’s project will use tools from information theory to explain when and why diffusion models succeed, identify their statistical limitations, and develop more efficient and privacy-preserving approaches. The research will also explore fundamental questions about generation quality, computational complexity, and the balance between accuracy and differential privacy. In addition to advancing the theory of modern machine learning, the project will provide opportunities for undergraduate and graduate students to train in cutting-edge statistical and computational methods.

Yuexi Wang
NSF Grant: New Methods for Scalable and Robust Simulation-Based Inference
Professor Yuexi Wang has received a National Science Foundation (NSF) grant to develop new methods for simulation-based inference (SBI), a powerful approach for analyzing complex scientific models across fields such as genetics, ecology, biology, economics, and psychology. Traditional statistical methods often struggle with these models because of their high-dimensional parameter spaces and imperfect alignment with real-world systems. This project will advance SBI by creating scalable techniques that can handle large, richly structured datasets and robust methods that account for model misspecification. By combining innovations such as score-based Langevin dynamics with new strategies to refine inference under imperfect models, Wang’s research aims to make simulation-based approaches more efficient, accurate, and broadly applicable. These advancements will provide scientists with stronger tools for understanding complex systems and improving decision-making across a wide range of disciplines.

Chan Park
NSF Grant: Collaborative Research: Distributional Balancing Methods for Advancing Causal Inference in Complex Settings
Professor Chan Park has been awarded a National Science Foundation (NSF) grant to advance causal inference methods for complex, real-world data. Many critical public health, education, and policy questions cannot be answered with randomized experiments, leaving researchers to rely on observational data that are often complicated by hidden biases, unmeasured factors, or interconnected influences among individuals. Park’s project focuses on improving distributional balancing methods, which create fair and comparable groups across observed variables, by extending them to account for clustered and networked data structures as well as unmeasured confounding. The research will also integrate instrumental variable techniques with advanced nonparametric approaches to strengthen causal conclusions. These methodological innovations will expand the statistical toolkit available to scientists and policymakers, leading to more reliable evidence for decision-making in areas such as healthcare, education, economics, and environmental policy.
These awards mark important milestones for each individual researcher and underscores the Department of Statistics’ commitment to advancing knowledge at the intersection of statistical theory, methodology, and real-world application.