BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//euroscipy-2024//talk//UNYV7V
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-euroscipy-2024-UNYV7V@pretalx.com
DTSTART;TZID=CET:20240826T140000
DTEND;TZID=CET:20240826T153000
DESCRIPTION:Data scientists are repeatedly told that it is absolutely criti
 cal to align their model training methodology with a specific business obj
 ective. While being a rather good advice\, it usually falls short on detai
 ls on how to achieve this in practice.\n\nThis hands-on tutorial aims to i
 ntroduce helpful theoretical concepts and concrete software tools to help 
 them bridge this gap. This method will be illustrated on a worked practica
 l use case: optimizing the operations of a fraud detection system for a pa
 yment processing platform.\n\nMore specifically\, we will introduce the co
 ncepts of calibrated probabilistic classifiers\, how to evaluate them and 
 fix common causes of mis-calibration. In a second part\, we will explore h
 ow to turn probabilistic classifiers into optimal business decision makers
 .\n\nThe tutorial material is available at the following URL: https://gith
 ub.com/probabl-ai/calibration-cost-sensitive-learning
DTSTAMP:20260308T034250Z
LOCATION:Room 5
SUMMARY:Probabilistic classification and cost-sensitive learning with sciki
 t-learn - Guillaume Lemaitre\, Olivier Grisel
URL:https://pretalx.com/euroscipy-2024/talk/UNYV7V/
END:VEVENT
END:VCALENDAR