BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Philipp Reichenbach (Technical University Berlin)
DTSTART:20200421T154000Z
DTEND:20200421T161000Z
DTSTAMP:20260423T005722Z
UID:NASO/8
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/NASO/8/">Inv
 ariant Theory and Matrix Normal Models</a>\nby Philipp Reichenbach (Techni
 cal University Berlin) as part of Max Planck Institute nonlinear algebra s
 eminar online\n\n\nAbstract\nWe describe connections between invariant the
 ory and maximum likelihood estimation (ML estimation)\, in the context of 
 matrix normal models. Namely\, we link ML estimation in that case to the l
 eft right action of SLxSL on tuples of matrices. This enables us to charac
 terize ML estimation by stability under that group action. Furthermore\, i
 nvariant theory provides a new upper bound on the sample size for generic 
 boundedness of the log-likelihood function. To illuminate the theory the t
 alk puts emphasis on several examples. At the end we briefly outline how o
 ur results generalize to Gaussian group models.\n\nBased on joint work wit
 h Carlos Améndola\, Kathlén Kohn and Anna Seigal. This is the second par
 t of a two part talk: in the first part\, Anna Seigal will discuss our res
 ults for log-linear models.\n
LOCATION:https://researchseminars.org/talk/NASO/8/
END:VEVENT
END:VCALENDAR