BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Computational Neuroscience Journal Club - Guillaume Hennequin (Uni
 versity of Cambridge)
DTSTART:20141007T150000Z
DTEND:20141007T160000Z
UID:TALK55139@talks.cam.ac.uk
CONTACT:Guillaume Hennequin
DESCRIPTION:Guillaume Hennequin will cover:\n\nInformation-limiting correl
 ations\n\nR. Moreno-Bote\, J. Beck\, I. Kanitscheider\, X. Pitkow\, P. Lat
 ham and A. Pouget\n\nNature Neuroscience\; vol. 17 (2014)\n\nhttp://www.na
 ture.com/neuro/journal/v17/n10/full/nn.3807.html\n\nABSTRACT :\n\nComputat
 ional strategies used by the brain strongly depend on the amount of inform
 ation that can be stored in population activity\, which in turn strongly d
 epends on the pattern of noise correlations. In vivo\, noise correlations 
 tend to be positive and proportional to the similarity in tuning propertie
 s. Such correlations are thought to limit information\, which has led to t
 he suggestion that decorrelation increases information. In contrast\, we f
 ound\, analytically and numerically\, that decorrelation does not imply an
  increase in information. Instead\, the only information-limiting correlat
 ions are what we refer to as differential correlations: correlations propo
 rtional to the product of the derivatives of the tuning curves. Unfortunat
 ely\, differential correlations are likely to be very small and buried und
 er correlations that do not limit information\, making them particularly d
 ifficult to detect. We found\, however\, that the effect of differential c
 orrelations on information can be detected with relatively simple decoders
 .
LOCATION:Cambridge University Engineering Department\, CBL\, BE-438 (http:
 //learning.eng.cam.ac.uk/Public/Directions)
END:VEVENT
END:VCALENDAR