BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Demixing scents: Sampling-based inference in olfaction. - Agnieszk
 a Grabska Barwinska\, The Gatsby Unit\, UCL
DTSTART:20130424T110000Z
DTEND:20130424T120000Z
UID:TALK44919@talks.cam.ac.uk
CONTACT:Dr. Cristina Savin
DESCRIPTION:Olfactory system faces similar problems to that of vision or a
 udition\, whereby a mixture of sources (visual objects\, voices or odours)
  causes a widespread activation of olfactory receptors. To infer what odou
 rs are impinging on the receptors is an over complete task\, since there a
 re many more odours than receptor types. The problem is exacerbated by the
  fact that odours rarely occur in solitude\, therefore the system needs to
  infer at least the major components of olfactory mixtures. To this end\, 
 one also needs to make a good guess on the (relative) concentrations of th
 e components. In my talk\, I will concentrate on an approximate scheme tha
 t follows this logic: our algorithm combines a Gibbs sampler that makes gu
 esses on odour components of the scene\, and a Langevin sampler that estim
 ates the posterior over concentrations in agreement with the current hypot
 hesis (on which odours are being present). We show how to avoid the proble
 m of synchronous updates in a network of neurons\, whose computations natu
 rally run in parallel. Finally\, the plausibility of our algorithm will be
  discussed and compared to an alternative inference scheme\, such as varia
 tional Bayes (Beck and all\, 2012).
LOCATION:Cambridge University Engineering Department\, CBL Rm #438 (http:/
 /learning.eng.cam.ac.uk/Public/Directions)
END:VEVENT
END:VCALENDAR