BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//programme.europython.eu//europython-2023//talk//SDEUCB
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-europython-2023-SDEUCB@programme.europython.eu
DTSTART;TZID=CET:20230720T103000
DTEND;TZID=CET:20230720T110000
DESCRIPTION:Don’t you love pickles? In the data science space\, the pickl
 e module has become one of the most popular ways to serialise and distribu
 te machine learning models - yet\, pickles introduce a wide range of probl
 ems. For starters\, it is incredibly easy to poison a pickle. Once this ha
 ppens\, a poisoned pickle can be used by an attacker to inject any arbitra
 ry code into your ML pipelines. And what’s even worse: it’s incredibly
  hard to detect if a pickle has been poisoned!\n\nGood news? Help is on th
 e way! You now have access to an increasing number of tools to help you ge
 nerate higher-quality pickles. And when those are not enough\, you can alw
 ays draw inspiration from the DevOps movement and their trust-or-discard p
 rocesses.\n\nThis talk will show you how widespread pickles are and how ea
 sy it is to poison models serialised with pickle\, but also how easy it is
  to start protecting them from attacks.
DTSTAMP:20260518T004804Z
LOCATION:South Hall 2A
SUMMARY:Poisoned pickles make you ill - Adrian Gonzalez-Martin
URL:https://programme.europython.eu/europython-2023/talk/SDEUCB/
END:VEVENT
END:VCALENDAR