BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2025//speaker//8SZGZW
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-pyconde-pydata-2025-3FUYVH@pretalx.com
DTSTART;TZID=CET:20250424T132500
DTEND;TZID=CET:20250424T141000
DESCRIPTION:"We've only tested the happy path - now users are finding all s
 orts of creative ways to break the app."\n\nWhat is already a cause for he
 adaches in traditional software engineering turns into a large challenge w
 hen the application is based on machine learning models: Data distribution
  may change from training phase to deployment. Even worse\, humans interac
 ting with the model may adjust their behaviour to the model making the gap
  between original training environment and deployment even larger. When de
 ployed in a public environment the model may be exposed to users trying to
  game the system. When re-trained it may be exposed to users trying to poi
 son the pool of training data.\n\nWe will take a tour of historic cases of
  models being gamed: What are the lessons we learnt a long time ago buildi
 ng e-mail spam filters? What happened when high search engine rankings sta
 rted to be linked to monetary income? How can personalization and targeted
  advertising be exploited to influence public discourse? \n\n“… it sho
 uld be clear that improvements in communication tend to divide mankind …
 ” by Harold Innis in Changing Concepts of Time\n\nThis keynote will turn
  interactive engaging the audience in sharing their stories on users playi
 ng interesting games with deployed models - including counter moves rolled
  out. \n\nIf we are to learn from IT security experience\, one important i
 ngredient to address these issues is a combination of collaboration and tr
 ansparency - across organisations.
DTSTAMP:20260423T023534Z
LOCATION:Zeiss Plenary (Spectrum)
SUMMARY:Machine Learning Models in a Dynamic Environment - Isabel Drost-Fro
 mm
URL:https://pretalx.com/pyconde-pydata-2025/talk/3FUYVH/
END:VEVENT
END:VCALENDAR