BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Some Applications of the Kullback-Leibler Divergence Rate in Hidde
 n Markov Models - Dr. M. Vidyasagar (Executive Vice President\, Tata Consu
 ltancy Services)
DTSTART:20080409T103000Z
DTEND:20080409T113000Z
UID:TALK11500@talks.cam.ac.uk
CONTACT:Dr Guy-Bart Stan
DESCRIPTION:The Kullback-Leibler (K-L) divergence rate between stochastic 
 processes is a generalization of the familiar K-L divergence between proba
 bility vectors.  In this talk\, the K-L divergence rate is introduced\, an
 d an easy derivation is given of the K-L divergence rate between two Marko
 v chains over a common alphabet.  This formula is used to solve the proble
 m of approximating a Markov chain with "long" memory by another with "shor
 t" memory in an optimal fashion.  The difficulties in extending this resul
 t to HMMs are explained.  Finally\, a geometrically convergent estimate fo
 r the K-L divergence rate between two HMMs is provided.
LOCATION: Cambridge University Engineering Department\, Lecture Room 6
END:VEVENT
END:VCALENDAR