Michael Kosorok

Michael R. Kosorok, PhD, is W. R. Kenan, Jr. Distinguished Professor of Biostatistics and Professor of Statistics and Operations Research at UNC-Chapel Hill.

His research expertise is in biostatistics, data science, machine learning and precision medicine, and he has written a major text on the theoretical foundations of these and related areas in biostatistics (Kosorok, 2008, Springer) as well as co-edited (with Erica E. M. Moodie, 2016, ASA-SIAM) a research monograph on dynamic treatment regimes and precision medicine.

He also has expertise in the application of biostatistics and data science to human health research, including cancer and cystic fibrosis. In particular, he is the contact principal investigator on an NCI program project grant (P01 CA142538), which focuses on statistical methods for novel cancer clinical trials in precision medicine, including biomarker discovery and dynamic treatment regimes. He has pioneered machine learning and data mining tools for these and related areas.

Invited Speakers

• Mouli Banerjee | University of Michigan
• Marie Davidian | North Carolina State University
• Peter Hoff | Duke University
• Jian Huang | Hong Kong Polytechnic University
• Yufeng Liu | University Of North Carolina Chapel Hill
• David Page | Duke University
• Chengchun Shi | London School of Economics and Political Science
• Stefan Wager | Stanford University
• Ming Yuan | Columbia University
• Donglin Zeng | Univeristy of Michigan

Call for posters!

We are excited to invite poster submissions for the upcoming Conference and Festschrift in Honor of Michael Kosorok, taking place on November 12-13, 2024, at the Carolina Club, UNC-Chapel Hill, George Watts Alumni Center. We welcome posters on any topics related to statistics, biostatistics, bioinformatics, machine learning, and artificial intelligence.

Submission Guidelines

Poster submissions are due on November 1st, 2024. Please submit your poster online at this Google Form Link with title, abstract (250-300 words) and author information. All poster presenters must also register for the conference (see conference information)

Poster Presentation

Posters will be displayed during the poster session on November 12, 2024, from 6:40 PM - 9:30 PM. Authors are required to be present at their posters during the session to discuss their work with conference attendees. Please note that posters must be printed, as digital posters will not be supported.

Awards

Posters will be judged by three distinguished statisticians attending the conference. An award will be presented to the most outstanding poster, which includes a $100 cash prize and a plaque signed and presented by Dr. Michael Kosorok.

Contact

For any questions, please contact Ruoqing Zhu (rqzhu@illinois.edu) and Nikki Freeman (nikki.freeman@duke.edu). We look forward to receiving your submissions and showcasing your research at the Conference in Honor of Michael Kosorok!

Conference Schedule

Tuesday, November 12th

5:00 PM - 5:15 PM Opening Remarks

Michael Hudgens, Professor and Chair, Department of Biostatistics, University of North Carolina at Chapel Hill

5:15 PM - 6:50 PM - Session I: Advances in Statistics

Chair: Ruoqing Zhu, University of Illinois Urbana-Champaign

• 5:15 PM - 5:45 PM - A Generalized Logrank-type Test for Comparison of Treatment Regimes in Sequential Multiple Assignment Randomized Trials

Marie Davidian (and Anastasios A Tsiatis), North Carlina State University

• 5:45 PM - 6:15 PM - Frequentist and Bayesian (FAB) Statistical Methods

Peter Hoff, Duke University

• 6:15 PM - 6:35 PM - Adaptive (Trainee) Learning

Tarek M Zikry, Columbia University

• 6:35 PM - 6:40 PM - Floor Discussion

6:40 PM - 9:30 PM - Banquet and Poster Session

Wednesday, November 13th

8:00 AM - 8:30 AM - Tea, Coffee, Breakfast

8:30 AM - 10:25 AM - Session II: Precision Medicine and Causal Inference

Chair: Yingqi Zhao, Fred Hutchinson Cancer Center

• 8:30 AM - 9:00 AM - Individualized Treatment Rules with Many Treatments

Yufeng Liu, University of North Carolina at Chapel Hill

• 9:00 AM - 9:20 AM - Independence Weights for Causal Inference with Continuous Treatments

Guanhua Chen, University of Wisconsin Madison

• 9:20 AM - 9:40 AM - Policy Learning with Continuous Actions Under Unmeasured Confounding

Ruoqing Zhu, University of Illinois Urbana-Champaign

• 9:40 AM - 10:00 AM - Heritability: a Counterfactual Perspective

Hongyuan Cao, Florida State University

• 10:00 AM - 10:20 AM - Robust Residual-weighted Learning for Optimal Treatment Regimes

Xin Zhou, Yale University

• 10:20 AM - 10:25 AM - Floor Discussion

10:25 AM - 10:35 AM - Break

10:35 AM - 12:30 PM - Session III: Experimental Design and Algorithmic Fairness

Chair: Guanhua Chen, University of Wisconsin Madison

• 10:35 AM - 11:05 AM - Ramblings on outcome weighted learning, direct search, and the paper that wouldn't die

Eric Laber, Duke University

• 11:05 AM - 11:35 PM - Switchback Experiments under Geometric Mixing

Stefan Wager, Stanford University

• 11:35 AM - 12:05 PM - Optimizing Individualized Treatments for a Target Population with Multiple Domains

Donglin Zeng, University of Michigan

• 12:05 PM - 12:25 PM - Improve Fairness of Neyman-Pearson Classifiers under Data Shift with Application to Early Detection of Cancer

Yingqi Zhao, Fred Hutchinson Cancer Center

• 12:25 PM - 12:30 PM - Floor Discussion

12:25 PM - 2:00 PM - Lunch

2:00 PM - 3:55 PM - Session IV: Machine Learning, Artificial Intelligence and High Dimensional Data

Chair: Nikki Freeman, Duke University

• 2:00 PM - 2:20 PM - LLM4Causal: Democratized Causal Tools for Everyone via Large Language Model

Rui Song, Amazon.com Inc

• 2:20 PM - 2:50 PM - Probably Approximately Correct Causal Discovery

David Page, Duke University

• 2:50 PM - 3:20 PM - Leveraging Large Models for Statistical Analysis: Two Case Studies from Generative Learning

Jian Huang, Hong Kong Polytechnic University

• 3:20 PM - 3:50 PM - Tensors in High Dimensional Data Analysis: Opportunities and Challenges

Ming Yuan, Columbia University

• 3:50 PM - 3:55 PM - Floor Discussion

3:55 PM - 4:10 PM - Break

4:10 PM - 5:45 PM - Session V: Censored Data and Empirical Process

Chair: Tarek M Zikry, Columbia University

• 4:10 PM - 4:40 PM - Dynamic Pricing Under Shape Constraints

Moulinath Banerjee, University of Michigan

We propose a shape-constrained approach to dynamic pricing for censored data in the linear valuation model that eliminates the need for tuning parameters commonly required in existing methods. Previous works have addressed the challenge of unknown market noise distribution F using strategies ranging from kernel methods to reinforcement learning algorithms, such as bandit techniques and upper confidence bounds (UCB). In contrast, we estimate F using isotonic regression which allows us to derive an upper bound on the seller's regret, comparable to UCB-based methods under similar assumptions. Empirical results from simulations and real-world data demonstrate that our method is competitive while offering the advantage of being completely tuning parameter-free. Time permitting, extensions of these ideas to s-concavity, a generalization of the log-concave shape constraint, and which arises naturally in dynamic pricing, will be discussed.

• 4:40 PM - 5:00 PM - Sequential Competing Risks Regression Model for Recurrent Event Data with Terminal Events

Anna Bellach, National Institute of Health

• 5:00 PM - 5:20 PM - Inference for Change-plane Regression

Chaeryon Kang, University of Pittsburgh

• 5:20 PM - 5:40 PM - Concentration Inequalities for Right-censored Data

Yair Goldberg, Technion - Israel Institute of Technology

• 5:40 PM - 5:45 PM - Floor Discussion

5:45 PM - 6:00 PM - Closing Remarks

Nancy Messonnier,
Dean and Bryson Distinguished Professor in Public Health,
Gillings School of Global Public Health,
University of North Carolina at Chapel Hill