BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:LCN Seminar: Stochastic computation in  spiking neural networks
DTSTART:20141204T133000
DTSTAMP:20260925T073443Z
UID:8064981b498f01c34745fbda5a4be986a1163bf6d64e693bd7959e30
CATEGORIES:Conferences - Seminars
DESCRIPTION:Mihai A. PETROVICI\; Electronic Vision(s) Group\, University o
 f Heidelberg\nWe know from experience that spiking neural networks are rem
 arkably efficient in solving complicated inference problems. Understanding
  these phenomena is not only interesting for brain science\, but also for 
 the development of novel\, neuro-inspired computing architectures. In this
  seminar\, we will review some recent approaches to spike-based probabilis
 tic computation that have been developed within the framework of the Brain
 ScaleS project. Starting from an abstract model of so-called neural sampli
 ng\, we will show how ensembles of leaky integrate-and-fire neurons can pr
 ovide a functional implementation of Boltzmann machines and Bayesian netwo
 rks. These theoretical considerations will be complemented by a discussion
  of interesting applications of spike-based inference\, with a particular 
 focus on the challenges and advantages of their implementation on mixed-si
 gnal\, accelerated neuromorphic hardware.
LOCATION:AAC132 http://plan.epfl.ch/?room=AAC132
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR