BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//juliacon-2026//talk//RGVXMP
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-juliacon-2026-RGVXMP@pretalx.com
DTSTART;TZID=CET:20260812T111500
DTEND;TZID=CET:20260812T113000
DESCRIPTION:The advent of convenient automatic differentiation has made gra
 dient-based optimization the default strategy for many challenging problem
 s and has revolutionized statistical practice.\nAt the same time\, MixedMo
 dels.jl uses a gradient-free approach to optimization and remains best in 
 class for linear mixed models.\nUsing MixedModels.jl as a case study\, we 
 will explore the tradeoffs of using the gradient and why gradient-free app
 roaches remain relevant even in a world of easy autodiff.
DTSTAMP:20260502T114917Z
LOCATION:Room 2
SUMMARY:Gradients aren't always great -- a case study with MixedModels.jl -
  Phillip Alday
URL:https://pretalx.com/juliacon-2026/talk/RGVXMP/
END:VEVENT
END:VCALENDAR