BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Dan Shiebler (Abnormal Security)
DTSTART:20240126T140000Z
DTEND:20240126T150000Z
DTSTAMP:20260423T021528Z
UID:ACPMS/33
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/ACPMS/33/">L
 earning with Kan Extensions</a>\nby Dan Shiebler (Abnormal Security) as pa
 rt of Algebraic and Combinatorial Perspectives in the Mathematical Science
 s\n\n\nAbstract\nA common problem in machine learning is "use this functio
 n defined over this small set to generate predictions over that larger set
 ." Extrapolation\, interpolation\, statistical inference and forecasting a
 ll reduce to this problem. The Kan extension is a powerful tool in categor
 y theory that generalizes this notion. In this work we explore application
 s of the Kan extension to machine learning problems. We begin by deriving 
 a simple classification algorithm as a Kan extension and experimenting wit
 h this algorithm on real data. Next\, we use the Kan extension to derive a
  procedure for learning clustering algorithms from labels and explore the 
 performance of this procedure on real data.\n
LOCATION:https://researchseminars.org/talk/ACPMS/33/
END:VEVENT
END:VCALENDAR