BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:www.tcs.tifr.res.in/event/1017
DTSTAMP:20230914T125947Z
SUMMARY:Bayesian Networks from a Reaction Theory Perspective
DESCRIPTION:Speaker: Manoj Gopalkrishnan (IIT\, Bombay)\n\nAbstract: \nAbst
ract:Bayesian networks are foundational objects in machine learning\, play
ing a central role as theoretical tools to understand and organize what we
know about machine learning\, including hierarchical Bayesian inference\,
belief propagation\, variational inference\, causality\, deep learning\,
etc. Reaction network theory is a well-developed and deep mathematical dis
cipline with a history dating back to the 1860s.\nWe show that every Bayes
ian networks can be described by a corresponding reaction network. This al
lows us to interpret biochemical reaction networks in living cells as perf
orming inference and learning\; to bring the tools of reaction network the
ory to the analysis of convergence problems in belief propagation\; to com
pare reaction network dynamics with approximate inference algorithms in ma
chine learning\; and to design reaction networks that are capable of perfo
rm machine learning in a solution.\nReferences:\n1. https://arxiv.org/abs/
1804.09062\n2. https://arxiv.org/abs/1906.09410\n
URL:https://www.tcs.tifr.res.in/web/events/1017
DTSTART;TZID=Asia/Kolkata:20191126T160000
DTEND;TZID=Asia/Kolkata:20191126T170000
LOCATION:A-201 (STCS Seminar Room)
END:VEVENT
END:VCALENDAR