BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Sparse Causal Learning
DTSTART:20240426T160000
DTEND:20240426T170000
DTSTAMP:20260407T115125Z
UID:ae520c744e724534baa0f7e67b626697422fb9136c082fb3259c8478
CATEGORIES:Conferences - Seminars
DESCRIPTION:Linbo Wang\, University  of Toronto\nAbstract: In many observ
 ational studies\, researchers are interested in studying the effects of mu
 ltiple exposures on the same outcome. Unmeasured confounding is a key chal
 lenge in these studies as it may bias the causal effect estimate. To mitig
 ate the confounding bias\, we introduce a novel device\, called the synthe
 tic instrument\, to leverage the information contained in multiple exposur
 es for causal effect identification and estimation. We show that under li
 near structural equation models\, the problem of causal effect estimation 
 can be formulated as an $\\ell_0$-penalization problem\, and hence can be 
 solved efficiently using off-the-shelf software. Simulations show that our
  approach outperforms state-of-art methods in both low-dimensional and hig
 h-dimensional settings. We further illustrate our method using a mouse obe
 sity dataset.\n \nBio: Linbo Wang is an assistant professor in the Depar
 tment of Statistical Sciences and the Department of Computer and Mathemati
 cal Sciences\, University of Toronto. He is also a faculty affiliate at th
 e Vector Institute\, a CANSSI Ontario STAGE program mentor\, and an Affili
 ate Assistant Professor in the Department of Statistics\, University of Wa
 shington\, and Department of Computer Science\, University of Toronto. Pri
 or to these roles\, he was a postdoc at Harvard T.H. Chan School of Public
  Health. He obtained his Ph.D. from the University of Washington. His rese
 arch interest is centered around causality and its interaction with statis
 tics and machine learning. \n\n 
LOCATION:CM 1 517 https://plan.epfl.ch/?room==CM%201%20517
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR