[Statlist] Research Seminar in Statistics | *FRIDAY 17 MAY 2024* | GSEM, University of Geneva

[gsem-support-instituts]

Dear All,

Last information for the next research seminar in Statistics which will be held on Friday 17 May 2024:

Access to the Uni Mail university building is permitted for anyone:
*       presenting a UNIGE multi-service card.
*       presenting a student card from HES-SO Genève or IHEID.
*       registered for a course, training, event, etc., upon presentation of proof of registration and a valid ID.

For the other participants: We invite you to register here: https://my.weezevent.com/rcs-seminar-17052024

Thank you for your attention.

Best regards,

Sandra


Sandra Vuadens
Assistant to research and institutes
gsem-support-instituts at unige.ch






-----Message d'origine-----
De : gsem-support-instituts
Envoyé : lundi, 13 mai 2024 09:30
Cc : Sebastian Engelke <Sebastian.Engelke at unige.ch>
Objet : Research Seminar in Statistics | *FRIDAY 17 MAY 2024* | GSEM, University of Geneva
Importance : Haute

Dear all,

We are pleased to invite you to our next Research Seminar, organized by Professor Sebastian Engelke on behalf of the Research Center for Statistics < https://www.unige.ch/gsem/en/research/institutes/rcs/team/ >.

FRIDAY 17 MAY 2024, at 11:15 am, Uni Mail M 4220.

A semiparametric perspective on unsupervised domain adaptation Jiwei ZHAO, University of Wisconsin-Madison, USA < https://biostat.wiscweb.wisc.edu/staff/zhao-jiwei/ >

ABSTRACT:
In studies ranging from clinical medicine to policy research, complete data are usually available from a population P, but the quantity of interest is often sought for a related but different population Q. In this talk, we consider the unsupervised domain adaptation setting under the label shift assumption. In the first part, we estimate a parameter of interest in population Q by leveraging information from P, where three ingredients are essential: (a) the common conditional distribution of X given Y, (b) the regression model of Y given X in P, and (c) the density ratio of the outcome Y between the two populations. We propose an estimation procedure that only needs some standard nonparametric technique to approximate the conditional expectations with respect to (a), while by no means needs an estimate or model for (b) or (c); i.e., doubly flexible to the model misspecifications of both (b) and (c). In the second part, we pay special attention to the case that the outcome Y is categorical. In this scenario, traditional label shift adaptation methods either suffer from large estimation errors or require cumbersome post-prediction calibrations. To address these issues, we propose a moment-matching framework for adapting the label shift, and an efficient label shift adaptation method where the adaptation weights can be estimated by solving linear systems. We rigorously study the theoretical properties of our proposed methods. Empirically, we illustrate our proposed methods in the MIMIC-III database as well as in some benchmark datasets including MNIST, CIFAR-10, and CIFAR-100.

BIOGRAPHY:
Jiwei Zhao is currently an Associate Professor (with tenure) at the University of Wisconsin-Madison, appointed by the Department of Biostatistics and Medical Informatics and affiliated with the Department of Statistics. His academic journey includes training at Nankai University where he was part of the Chern Honored Class of Mathematics, and obtaining his PhD in Statistics from UW-Madison. He embarked on his academic career as an Assistant Professor at the State University of New York at Buffalo in 2014, and was promoted to Associate Professor (with tenure) in 2020. While at Buffalo, he was also honored with the UB Exceptional Scholar - Young Investigator Award in 2020. Moreover, his accomplishments led to his election as an Elected Member of the International Statistical Institute in 2021. His statistical areas of interest include semiparametric statistics, tradeoff between efficiency and robustness, missing data analysis and causal inference, high-dimensional data analysis and high-dimensional statistical inference, domain adaptation and transfer learning. His domain areas of interest include patient-reported outcome, clinical trial, real world evidence and real world data, health disparity and health equity, and health services research.

> View the Research Seminar agenda: <
> https://www.unige.ch/gsem/en/research/seminars/rcs/ >

Regards,


Marie-Madeleine

Marie-Madeleine Novo
Assistant to the Research Institutes
gsem-support-instituts at unige.ch