With the availability of data and information from multiple sources and domains, the past two decades have witnessed an explosive evolution on statistical methodologies for model transportability, generalizability, data exploitation, integration, and fusion. Concurrently, the machine learning community has also forged ahead with the creation of algorithms and approaches for transfer learning, out-of-distribution prediction, semi-supervised learning, and federated learning. While many research concepts and proposals share a pertinent and akin nature, they have not yet garnered unanimous recognition.
A primary objective of this workshop is to convene statisticians, biostatisticians, computer scientists, epidemiologists, social scientists, and industry researchers. The aim is twofold: firstly, to showcase the most recent advancements in this expansive realm of research; and secondly, to cast a visionary gaze towards new horizons that lie ahead.
Poster Session
This workshop will include a poster session for early career researchers (including graduate students). In order to propose a poster, you must first register for the workshop, and then submit a proposal using the form that will become available on this page after you register. The registration form should not be used to propose a poster.
The deadline for proposing is Sunday, April 12, 2026. If your proposal is accepted, you should plan to attend the event in-person.
Funding
There is limited funding for participants, and funding is not guaranteed. Funding requests received by Sunday, April 12, 2026 will have priority.
In-Person Registration
Seats are limited at the venue, which means that in-person registration may be capped prior to the workshop start date. If capacity is reached, a waitlist will be imposed, which the registration form will reflect. Early registration is strongly encouraged.
All in-person registrants must wait to receive an invitation to attend in-person from IMSI before traveling, which generally begin to be sent out 4-6 weeks in advance.
All registrants (online and in-person) will receive zoom links and are welcome to attend online.
Sijia Li
University of California, Los Angeles (UCLA)
X
L
Xiudi Li
University of California, Berkeley
K
M
Krikamol Muandet
CISPA Helmholtz Center for Information Security
H
N
Hongseok Namkoong
Columbia University
Y
N
Yang Ning
Cornell University
H
P
Harsh Parikh
Yale University
J
S
Jiawei Shan
University of Wisconsin
P
S
Peter Song
University of Michigan
E
S
Elizabeth Stuart
Johns Hopkins University
J
W
Jindong Wang
College of William and Mary
Y
W
Ying Wei
Columbia University
S
Y
Shu Yang
North Carolina State University
H
Z
Heping Zhang
Yale University
T
Z
Tong Zhang
University of Illinois Urbana-Champaign
Schedule
Monday, June 22, 2026
8:30-9:00 CDT
Breakfast/Check-in
9:00-9:30 CDT
Introduction/Logistics/Housekeeping
9:30-10:10 CDT
Estimating Long-Term Treatment Effects by Integrating Randomized Trials and Real-World Data under Unmeasured Confounding, Study Heterogeneity, and Informative Dropout
Speaker: Shu Yang (North Carolina State University)
10:10-10:15 CDT
Q&A
10:15-10:20 CDT
Tech Break
10:20-11:00 CDT
Transfer learning for ridge regression with random coefficients
Speaker: Hongzhe Li (University of Pennsylvania)
11:00-11:05 CDT
Q&A
11:05-11:35 CDT
Coffee Break
11:35-12:15 CDT
Unsupervised Federated Multi-Task Learning for Heterogeneous Tasks
Speaker: Yang Feng (New York University)
12:15-12:20 CDT
Q&A
12:20-13:20 CDT
Lunch Break
13:20-14:00 CDT
Active Subsampling for Binary Response Models
Speaker: Yang Ning (Cornell University)
14:00-14:05 CDT
Q&A
14:05-14:10 CDT
Tech Break
14:10-14:50 CDT
Bregman Projection for Calibration Estimation
Speaker: Jae-Kwang Kim (Iowa State University)
14:50-14:55 CDT
Q&A
14:55-15:25 CDT
Coffee Break
15:25-16:05 CDT
Semiparametric Mediation Analysis with Separately Observed Mediator and Outcome under Unmeasured Confounding
Speaker: Sijia Li (University of California, Los Angeles (UCLA))
16:05-16:10 CDT
Q&A
Tuesday, June 23, 2026
8:30-9:00 CDT
Breakfast/Check-in
9:00-9:40 CDT
Improving Personalization and Consistency of Large Foundation Models
Speaker: Jindong Wang (College of William and Mary)
9:40-9:45 CDT
Q&A
9:45-9:50 CDT
Tech Break
9:50-10:30 CDT
Learning under Partial Knowledge: Robust Generalisation and Imprecise Forecasting
Speaker: Krikamol Muandet (CISPA Helmholtz Center for Information Security)
As machine learning systems are deployed in increasingly open-ended and high-stakes environments, their fundamental limitations become more apparent. Distribution shift, adversarial manipulation, catastrophic forgetting, hallucinations, safety failures, and misalignment all stem from a common challenge: making decisions at the boundary between what a model knows and what it does not know. Yet most contemporary learning systems are designed to produce precise predictions even when the available information is incomplete or unreliable.
In this talk, I will present recent work on robust out-of-distribution generalisation and imprecise forecasting that seeks to address this challenge. The unifying theme is learning under partial knowledge: developing intelligent systems that can explicitly represent, reason about, and communicate the limits of their own knowledge. Rather than treating ignorance as a hidden vulnerability, these approaches elevate it to a first-class object of inference, enabling models to make more robust predictions and more trustworthy decisions under uncertainty.
10:30-10:35 CDT
Q&A
10:35-11:05 CDT
Coffee Break
11:05-11:45 CDT
Variance Reduction and Adaptive Sampling in Reinforcement Learning with Verifiable Rewards
Speaker: Tong Zhang (University of Illinois Urbana-Champaign)
11:45-11:50 CDT
Q&A
11:50-11:55 CDT
Tech Break
11:55-12:35 CDT
Learning Surrogate Indices from Historical A/Bs: Adversarial ML for Debiased Inference on Functionals of Ill-Posed Inverses
Speaker: Nathan Kallus (Cornell University)
12:35-12:40 CDT
Q&A
12:40-13:40 CDT
Lunch Break
13:40-15:10 CDT
Lightning Talks Session One
15:10-16:30 CDT
Poster Session and Social Hour
Wednesday, June 24, 2026
8:30-9:00 CDT
Breakfast/Check-in
9:00-9:40 CDT
TBA
Speaker: Heping Zhang (Yale University)
9:40-9:45 CDT
Q&A
9:45-9:50 CDT
Tech Break
9:50-10:30 CDT
Integrating Epigenetic Clocks via Deep Neural Networks
Speaker: Peter Song (University of Michigan)
10:30-10:35 CDT
Q&A
10:35-11:05 CDT
Coffee Break
11:05-11:45 CDT
The Promise and Pitfalls of Integrating Data for Causal Inference, With Application to Mental Health Treatment
Speaker: Liz Stuart (University of Michigan)
11:45-11:50 CDT
Q&A
11:50-11:55 CDT
Tech Break
11:55-12:35 CDT
Targeted Data Fusion for Region-Specific Survival Effects in the AMP HIV Prevention Trials
Speaker: Larry Han (Northeastern University)
12:35-12:40 CDT
Q&A
12:40-13:40 CDT
Lunch Break
13:40-15:10 CDT
Panel: National Academies Report on “Frontiers of Statistics in Science and Engineering: 2035 and Beyond”
Panelists: Lance Waller (Emory University), Tian Zheng (Columbia University), Amy Braverman (Jet Propulsion Laboratory, California Institute of Technology), and Frauke Kreuter (University of Maryland)
15:10-15:40 CDT
Coffee Break
15:40-16:20 CDT
Efficient Inference for Time-to-Event Outcomes by Integrating Right-Censored and Current Status Data
Speaker: Xiudi Li (University of California, Berkeley)
16:20-16:25 CDT
Q&A
Thursday, June 25, 2026
8:30-9:00 CDT
Breakfast/Check-in
9:00-9:40 CDT
Causal Invariance Learning via Efficient Nonconvex Optimization
Speaker: Zijian Guo (Zhejiang University (ZJU))
Identifying causal relationships from observational data is a fundamental yet challenging problem. This paper focuses on learning the causal outcome model, namely identifying the direct causes of an outcome and estimating their effects. We leverage data collected from multiple heterogeneous environments, which enable causal discovery through the invariance principle that the causal outcome model remains invariant across environments. Based on this principle, we propose Negative Weighted Distributionally Robust Optimization (NegDRO), a framework that minimizes the worst-case weighted combination of risks across environments and enforces invariance by allowing negative weights.
Under the additive interventions regime, we establish three main contributions. From a statistical perspective, we provide sufficient and nearly necessary identification conditions under which the NegDRO solution coincides with the true causal outcome model. From an optimization perspective, despite the inherent nonconvexity of the NegDRO objective, we show that it admits a benign optimization landscape in which all stationary points lie close to the causal outcome model. From a computational perspective, we develop a gradient-based algorithm that provably converges to the causal outcome model with non-asymptotic rates in both sample size and number of iterations.
In particular, NegDRO avoids exhaustive combinatorial searches over subsets of covariates used in existing approaches and scales efficiently to high-dimensional settings. To our knowledge, this is the first causal invariance learning method that solves a nonconvex optimization problem to global optimality.
9:40-9:45 CDT
Q&A
9:45-9:50 CDT
Tech Break
9:50-10:30 CDT
Designing Multi-Site Studies for External Validity: Site Selection via Synthetic Purposive Sampling
Speaker: Naoki Egami (Massachusetts Institute of Technology (MIT))
10:30-10:35 CDT
Q&A
10:35-11:05 CDT
Coffee Break
11:05-11:45 CDT
Adaptive Experimentation via Autoregressive Generation
Speaker: Hongseok Namkoong (University of Pennsylvania)
11:45-11:50 CDT
Q&A
11:50-11:55 CDT
Tech Break
11:55-12:35 CDT
TBA
Speaker: Ying Wei (Columbia University)
12:35-13:40 CDT
Lunch Break
13:40-15:10 CDT
Lightning Talks Session Two
15:10-16:30 CDT
Poster Session Two and Social Hour
Friday, June 26, 2026
8:30-9:00 CDT
Breakfast/Check-in
9:00-9:40 CDT
Transporting Causal Effects Beyond Common Support
Speaker: Harsh Parikh (Yale University)
9:40-9:45 CDT
Q&A
9:45-10:15 CDT
Coffee Break
10:15-10:55 CDT
Beyond Exchangeability: Distribution-Shift-Aware Integration of External Control Data in Randomized Trials
Speaker: Jiawei Shan (University of Wisconsin)
10:55-11:00 CDT
Q&A
11:00-11:30 CDT
Closing Remarks and Workshop Survey
Registration
IMSI is committed to making all of our programs and events inclusive and accessible.
Contact [email protected] to request
disability-related accommodations.
In order to register for this workshop, you must have an IMSI account and be logged in.