User Tools

Site Tools


seminars:stat:mar282024

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
seminars:stat:mar282024 [2024/03/19 10:51] qyuseminars:stat:mar282024 [2024/03/19 10:54] (current) qyu
Line 1: Line 1:
 +<WRAP centeralign>##Statistics Seminar##\\ Department of Mathematics and Statistics</WRAP>
 +
 +<WRAP 70% center>
 +^  **DATE:**|Thursday, March 28, 2024 |
 +^  **TIME:**|1:15pm -- 2:15pm |
 +^  **LOCATION:**|WH 100E |
 +^  **SPEAKER:**|Baozhen Wang, Binghamton  University |
 +^  **TITLE:**|Conformal Meta-learners for Predictive Inference of Individual Treatment Effects  |
 +</WRAP>
 +\\ 
 +
 +<WRAP center box 80%>
 +<WRAP centeralign>**Abstract**</WRAP>
 +This paper investigates predictive inference for individual
 +treatment effects (ITEs) using machine learning techniques. Traditional
 +approaches have primarily concentrated on developing meta-learners for
 +estimating the conditional average treatment effect (CATE), offering point
 +estimates without considering predictive intervals. The study introduces
 +conformal meta-learners, a framework that enhances traditional CATE
 +meta-learners by applying the conformal prediction procedure to provide
 +predictive intervals for ITEs. This method is validated through a
 +stochastic ordering framework, highlighting that conformal meta-learners
 +can achieve valid inferences with desired coverage levels.
 +</WRAP>
 +
 +
 +
 +