BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian inference with probabilistic population codes - Alexandre
  Pouget\, University of Rochester
DTSTART:20060321T130000Z
DTEND:20060321T140000Z
UID:TALK4754@talks.cam.ac.uk
CONTACT:Cordula Becker
DESCRIPTION:Recent psychophysical experiments indicate that humans perform
  near-optimal Bayesian inference in a wide variety of tasks\, ranging from
  cue integration to decision-making to motor control. This implies that ne
 urons both represent probability distributions and combine those distribut
 ions according to a close approximation to Bayes rule. At first sight\, it
  would appear that the high variability in the responses of cortical neuro
 ns would make it difficult to implement such optimal statistical inference
  in cortical circuits. I will show that\, in fact\, this variability gener
 ates probabilistic population codes which represent probability distributi
 ons over the encoded stimulus. Moreover\, when the neural variability is P
 oisson-like\, as is the case in cortex\, a broad class of Bayesian inferen
 ce\, such as cue integration\, or integrating evidence for decision-making
  \, can be closely approximated with simple linear combinations of probabi
 listic population codes. Therefore\, this theory suggests that the Poisson
 -like variability in the cortex greatly simplifies Bayesian inference in n
 eural circuits. 
LOCATION:Seminar Room (ground floor)\, Craik-Marshall Building
END:VEVENT
END:VCALENDAR