Workshop: quantifying structure in large neural datasets
Columbia University, Oct 16-18, 2013
Due to the significant interest in this event, we
have had to close registration to avoid overflowing our
room.
In systems neuroscience, it is becoming increasingly common to record
the activity of hundreds of neurons simultaneously via electrode
arrays or optical imaging. Emerging technologies promise further
increases in the number of simultaneous recordings. We will thus
often be limited not by our measurements but by our ability to
interpret those measurements. How should one translate a long list of
spike times / activity levels into an understanding of what the neural
population is doing? Is a specialized approach always necessary or
can we agree on a set of standardized techniques? What quantitative
tools are currently available, and what have they taught us so far?
What tools do we need that have yet to be developed? The goal of this
meeting is to propagate a dialogue that focuses on such questions.
Tentative Schedule:
October 16th
8:00 Continental Breakfast
9:00 Anne Churchland, Cold Spring Harbor: Linking circuits to behavior in visual decision making
9:40 Misha Ahrens, Janelia Farm Research Campus: Motor learning and whole-brain imaging in zebrafish
10:20 Coffee
10:40 Andreas Tolias, Baylor College of Medicine: The organization of the neocortical microcircuit
11:20 Matthias Bethge, Max Planck Institute for Biological Cybernetics: Beyond GLMs: A generative mixture modeling approach to neural system identification
12:00 Lunch
1:20 Jonathan Pillow, UT Austin: Scalable Bayesian nonparametric models for binary spike patterns
2:00 Ryan Adams, Harvard University: Latent variable extensions to the generalized linear model
2:40 Lawrence Carin, Duke University: Sorting electrophysiological data via dictionary learning & mixture modeling
3:20 Coffee
3:40 Christian Machens, Champalimaud Center for the Unknown: Statistical commonalities in population responses across brain areas
4:20 Matthew Harrison, Brown University: Robust inference for nonstationary spike trains
October 17th
8:00 Continental Breakfast
9:00 Byron Yu, Carnegie Mellon University: Neural constraints on learning
9:40 Diego Gutnisky, Janelia Farm Research Campus: Population coding in an active somatosensation task
10:20 Coffee
10:40 Paul Cisek, University of Montreal: Inferring the mechanisms of decisions
11:20 John Cunningham, Columbia University: Model testing using neural populations
12:00 Lunch
1:30 William Newsome, Stanford University: Targeted dimensionality reduction
2:10 David Sussillo, Stanford University: A neural network that finds naturalistic solutions for the production of muscle activity
2:50 Coffee
3:20 Steve Scott, Queen's University: Task dependent feedback processing in sensory and motor cortices
4:00 Maneesh Sahani, Gatsby Computational Unit: Inferring population dynamics from ensemble cortical activity
October 18th
8:00 Continental Breakfast
9:00 Ken Harris, Imperial College London: The population rate
9:40 Matteo Carandini, University College London: The nature of cortical noise
10:20 Coffee
10:40 Rob Kass, Carnegie Mellon University: Some statistical considerations in making inferences about neural networks
11:20 Carlos Brody, Princeton University: Prospects for cellular-resolution multi-neuron imaging of executive control processes
12:00 Closing remarks
All talks will be in the Presidential Ballroom, on the third floor of
the Faculty House.
Note that the address is technically 64 Morningside drive. However,
one must enter from a large gate on the north side of 116th Street
between Amsterdam and Morningside. A map from the 1 subway train to the
Faculty House is here.
« Back to the Grossman Center for the Statistics of Mind main page