BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:0209efb182d4a7b45449eae9e7ab2ebe
DTSTAMP:20260404T054341Z
SUMMARY:Talk by Ricardo Silva
DESCRIPTION:Title: Some machine learning tools to aid causal inference\nAbs
 tract: Causal inference from observational data requires untestable assump
 tions. As assumptions may fail\, it is important to be able to understand 
 how conclusions vary under different premises.  Machine learning methods a
 re particularly good at searching for hypotheses\, but they do not always 
 provide ways of expressing a continuum of assumptions from which causal es
 timands can be proposed.  We introduce one family of assumptions and algor
 ithms that can be used to provide alternative explanations for treatment e
 ffects.  If we have time\, I will also discuss some other developments on 
 the integration of observational and interventional data using a nonparame
 tric Bayesian approach.
URL:https://www.imperial.ac.uk/events/100350/talk-by-ricardo-silva/
DTSTART;TZID=Europe/London:20170623T140000
DTEND;TZID=Europe/London:20170623T150000
LOCATION:United Kingdom
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
DTSTART:20170623T140000
TZNAME:BST
TZOFFSETTO:+0100
TZOFFSETFROM:+0100
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR