BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2025//talk//JH97CL
BEGIN:VTIMEZONE
TZID:Europe/Berlin
BEGIN:DAYLIGHT
DTSTART:20240425T000000
TZNAME:CEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0200
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20241027T030000
RDATE:20251026T030000
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250330T030000
RDATE:20260329T030000
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:Pipeline-level differentiable programming for the real world - Ale
 ssandro Angioi
DTSTART;TZID=Europe/Berlin:20250425T140000
DTEND;TZID=Europe/Berlin:20250425T143000
DTSTAMP:20260811T164351Z
UID:pretalx-pyconde-pydata-2025-JH97CL@pretalx.com
DESCRIPTION:Automatic Differentiation (AD) is not only the backbone of mod
 ern deep learning but also a transformative tool across various domains su
 ch as control systems\, materials science\, weather prediction\, 3D render
 ing\, data-driven scientific discovery\, and so on. Thanks to a mature ML 
 framework ecosystem\, powered by libraries like PyTorch and JAX\, AD perfo
 rms remarkably well at a component level\; however\, integrating these com
 ponents into differentiable pipelines still remains a significant challeng
 e. In this talk\, we will provide an accessible introduction to (pipeline-
 level) AD\, demonstrate some cool applications you can build with it\, and
  see how to build differentiable pipelines that hold up in the real world.
LOCATION:Hassium
URL:https://pretalx.com/pyconde-pydata-2025/talk/JH97CL/
END:VEVENT
END:VCALENDAR