BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//AMSE//Event Calendar//FR
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:event-11375@www.amse-aixmarseille.fr
DTSTAMP:20260422T181800Z
CREATED:20260422T181800Z
LAST-MODIFIED:20260422T181800Z
STATUS:CONFIRMED
SEQUENCE:0
SUMMARY:amse seminar - Kosuke Imai
DTSTART:20241014T093000Z
DTEND:20241014T104500Z
DESCRIPTION:The use of Artificial Intelligence (AI)\, or more generally dat
 a-driven algorithms\, has become ubiquitous in today's society. Yet\, in ma
 ny cases and especially when stakes are high\, humans still make final dec
 isions. The critical question\, therefore\, is whether AI helps humans mak
 e better decisions compared to a human-alone or AI-alone system. We intro
 duce a new methodological framework to empirically answer this question wit
 h a minimal set of assumptions. We measure a decision maker's ability to m
 ake correct decisions using standard classification metrics based on the ba
 seline potential outcome. We consider a single-blinded and unconfounded t
 reatment assignment\, where the provision of AI-generated recommendations 
 is randomized across cases with humans making final decisions. Under this 
 study design\, we show how to compare the performance of three alternative
  decision-making systems — human-alone\, human-with-AI\, and AI-alone. I
 mportantly\, the AI-alone system includes any individualized treatment ass
 ignment\, including those that are not used in the study. We also show whe
 n to provide a human-decision maker with AI recommendations and when they 
 should follow such recommendations. We apply the proposed methodology to 
 the data from our own randomized controlled trial of a pretrial risk assess
 ment instrument. We find that the risk assessment recommendations do not 
 improve the classification accuracy of a judge's decision to impose cash b
 ail. Our analysis also shows that both the risk assessment score and a lar
 ge language model generally perform worse than human decisions.  \\n\\nCo
 ntact: Nicolas Clootens: nicolas.clootens[at]univ-amu.frRomain Ferrali: rom
 ain.ferrali[at]univ-amu.fr\n\nPlus d'informations: https://www.amse-aixmars
 eille.fr/en/events/kosuke-imai-0
LOCATION:Îlot Bernard du Bois - Amphithéâtre\, AMU - AMSE\, 5-9 boulevar
 d Maurice Bourdet\, 13001 Marseille
URL;VALUE=URI:https://www.amse-aixmarseille.fr/en/events/kosuke-imai-0
CONTACT:Nicolas Clootens: nicolas.clootens[at]univ-amu.frRomain Ferrali: ro
 main.ferrali[at]univ-amu.fr
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR
