BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Signal recovery in cortico-cortical communication through represen
 tational learning in the brain
DTSTART:20101028T161500
DTSTAMP:20260429T010421Z
UID:6af3f4c19a01c4e769ebaa876342e26900374b306cbaaeec87e32b25
CATEGORIES:Conferences - Seminars
DESCRIPTION:Friedrich T. SOMMER\nThe talk will discuss different objective
 s of representational learning in the brain by describing the results from
  two studies. The first study investigates if the objective of efficient c
 oding can explain the shape diversity of receptive fields found in V1 (Rin
 gach\, 2002). Early models of efficient sparse coding (e.g. Olshausen & Fi
 eld 1996) were unable to reproduce the full diversity of shapes observed i
 n V1. We developed network models for V1 that differed in two respects fro
 m the early models. First\, the learned representations are overcomplete\,
  that is\, the cortical neurons in the model outnumber the dimension of th
 e thalamic input. Second\, the sparseness constraint forces the number of 
 active neurons to be small\, in contrast to limiting the average neuronal 
 activity as done in the earlier models. The new model accurately predicted
  the distribution of shapes of cortical receptive fields found in nature s
 uggesting that efficient coding is crucial\, however\, the high dimensiona
 lity and the type of sparseness are other critical features of the model (
 Rehn & Sommer\, 2007). \nThe second study takes off where the first ends. 
 How can high-dimensional sparse representations formed in V1 be transmitte
 d through long-range fiber projections to other brain regions? The problem
  is that the number of axons sent out from one region to another is much s
 maller than the number of local neurons in each region (Schuez et al. Cere
 bral Cortex\, 2006). Combining ideas from sparse coding and compressive sa
 mpling\, we have discovered a learning algorithm\, adaptive compressive sa
 mpling\, that can be proven to establish and maintain lossless communicati
 on through fiber bottlenecks. Our discovery can explain how a neural popul
 ation in the brain targeted by an axonal fiber projection can use the arri
 ving signals to learn response properties that not only convey the full in
 formation sent into the projection but also resemble experimentally observ
 ed receptive fields. Thus\, we argue\, the critical objective of represent
 ational learning might be the recovery of subsampled signals\, a hard and 
 ubiquitous problem for long-range communication in the brain. Efficient co
 ding emerges as a necessary byproduct (Isely et al. NIPS 2010).
LOCATION:BC 02 https://plan.epfl.ch/?room==BC%2002
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR