BEGIN:VCALENDAR
VERSION:2.0
PRODID:talks.ox.ac.uk
BEGIN:VEVENT
SUMMARY:Simplicity and Complexity of Belief-Propagation - Professor Elchan
an Mossel (MIT Statistics + Data Center)
DTSTART;VALUE=DATE-TIME:20191017T160000
DTEND;VALUE=DATE-TIME:20191017T170000
UID:https://talks.ox.ac.uk/talks/id/e99818c0-f8cb-4a06-89ca-e539fd6d589d/
DESCRIPTION:T\\here is a very simple algorithm for the inference of poster
iors for probability models on trees. This algorithm\, known as “Belief
Propagation” is widely used in coding theory\, in machine learning\, in
evolutionary inference\, among many other areas. The talk will be devoted
to the analysis of Belief Propagation in some of the simplest probability
models. We will highlight the interplay between Belief Propagation\, linea
r estimators (statistics)\, the Kesten-Stigum bound (probability) and Repl
ica Symmetry Breaking (statistical physics). We will show how the analysis
of Belief Propagation allowed to proof phase transitions for phylogenetic
reconstruction in evolutionary biology and develop optimal algorithms for
inference of block models. Finally\, we will discuss the computational co
mplexity of this “simple” algorithm.\nSpeakers:\nProfessor Elchanan Mo
ssel (MIT Statistics + Data Center)
LOCATION:24-29 St Giles' (Large Lecture Theatre\, Department of Statistics
)\, 24-29 St Giles' OX1 3LB
TZID:Europe/London
URL:https://talks.ox.ac.uk/talks/id/e99818c0-f8cb-4a06-89ca-e539fd6d589d/
BEGIN:VALARM
ACTION:display
DESCRIPTION:Talk:Simplicity and Complexity of Belief-Propagation - Profess
or Elchanan Mossel (MIT Statistics + Data Center)
TRIGGER:-PT1H
END:VALARM
END:VEVENT
END:VCALENDAR