BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Using ConvNets\, MALIS and crowd-sourcing to map the retinal conne
 ctome - Srini Turaga\, Gatsby Unit &amp\; Wolfson Institute for Biomedical
  Research\, UCL
DTSTART:20140319T110000Z
DTEND:20140319T120000Z
UID:TALK51466@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:Neural circuits in the brain are formed from neurons connectin
 g to one another in highly structured ways. However\, technological limita
 tions have prevented us from knowing much about the nature of neural conne
 ctivity and how it relates to neural computation. We have developed new te
 chnology based on 3d electron microscopy\, computational image analysis an
 d crowd-sourcing to reconstruct complete wiring diagrams for all the neuro
 ns in a piece of brain tissue.\n\nWe have densely reconstructed 950 neuron
 s in the inner plexiform layer of the mouse retina using a combination of 
 machine learning algorithms and human proof-reading. These reconstructions
  yield hints of the principles underlying neural connectivity and neural c
 omputation in the retina. I will briefly describe these results and presen
 t the computational methods leading to this work. Our machine learning met
 hod for image segmentation is a deep convolutional neural network (ConvNet
 )\, which when combined with the novel global cost function for image segm
 entation (MALIS) yields neuron tracing accuracy approaching that of a sing
 le human expert (tracings from multiple human experts are usually combined
  and proofread to increase tracing accuracy).\n\nJoint work with Viren Jai
 n\, Moritz Helmstaedter\, Kevin Briggman\, Winfried Denk and Sebastian Seu
 ng.\n
LOCATION:Auditorium\, Microsoft Research Ltd\, 21 Station Road\, Cambridge
 \, CB1 2FB
END:VEVENT
END:VCALENDAR