BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:From Gossip to Voting
DTSTART:20140617T151500
DTSTAMP:20260407T002723Z
UID:3aed91d2a79baa956208fc43ca88f3bf94a02b9995f25d19a994691e
CATEGORIES:Conferences - Seminars
DESCRIPTION:Patrick Thiran\, EPFL\nAn increasingly larger number of applic
 ations require networks to perform decentralized computations over distrib
 uted data. A representative problem of these ``in-network processing" task
 s is the distributed computation of the average of values present at nodes
  of a network\, known as gossip algorithms. They have received recently si
 gnificant attention across different communities (networking\, algorithms\
 , signal processing\, control) because they constitute simple and robust m
 ethods for distributed information processing over networks. The first par
 t of the talk is a short survey of some results on real-valued (analog) go
 ssip algorithms. The second part is devoted to quantized gossip on arbitra
 ry connected networks\, and to a particular instance of this problem\, the
  voting problem: nodes initially vote for Yes (1) or No (0)\, and they wan
 t to know the majority opinion. We show that the majority voting problem i
 s solvable with only 2 bits of memory per agent.\n(This is a joint work wi
 th Florence Bénézit and Martin Vetterli).
LOCATION:BC 420 https://plan.epfl.ch/?room==BC%20420
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR