BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2024//talk//YPKKQF
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-2024-YPKKQF@pretalx.com
DTSTART;TZID=CET:20240423T163500
DTEND;TZID=CET:20240423T170500
DESCRIPTION:In the last year there hasn’t been a day that passed without 
 us hearing about a new generative AI innovation that will enhance some asp
 ect of our lives. On a number of tasks large probabilistic systems are now
  outperforming humans\, or at least they do so “on average”. “On ave
 rage” means most of the time\, but in many real life scenarios “averag
 e” performance is not enough: we need correctness ALL of the time\, for 
 example when you ask the system to dial 911. \n\nIn this talk we will expl
 ore the synergy between deterministic and probabilistic models to enhance 
 the robustness and controllability of machine learning systems. Tailored f
 or ML engineers\, data scientists\, and researchers\, the presentation del
 ves into the necessity of using both deterministic algorithms and probabil
 istic model types across various ML systems\, from straightforward classif
 ication to advanced Generative AI models. \n\nYou will learn about the uni
 que advantages each paradigm offers and gain insights into how to most eff
 ectively combine them for optimal performance in real-world applications. 
 I will walk you through my past and current experiences in working with si
 mple and complex NLP models\, and show you what kind of pitfalls\, shortcu
 ts\, and tricks are possible to deliver models that are both competent and
  reliable.\n\nThe session will be structured into a brief introduction to 
 both model types\, followed by case studies in classification and generati
 ve AI\, concluding with a Q&A segment.
DTSTAMP:20260515T075251Z
LOCATION:B07-B08
SUMMARY:Would you rely on ChatGPT to dial 911? A talk on balancing determin
 ism and probabilism in production machine learning systems - Nicolas Gueno
 n des Mesnards
URL:https://pretalx.com/pyconde-pydata-2024/talk/YPKKQF/
END:VEVENT
END:VCALENDAR