[Ml-stat-talks] Fwd: [pni-all] Tomorrow: January 30th - PNI Theory Symposium - A32 Princeton Neuroscience Institute

David Blei blei at CS.Princeton.EDU
Wed Jan 29 15:57:10 EST 2014


these announcements come from ken norman in neuroscience.  enjoy the fancy
new building while hearing great talks!

best
dave








[image: Final_PNI_horiz_logo]



Theory Symposium

Thursday, January 30th

A32 Princeton Neuroscience Institute

9:30am-12:00pm




*9:30am*

*Odelia Schwartz*

*Albert Einstein College of Medicine *

*"Understanding contextual visual processing using a principled model of
natural image statistics"*



Neural processing, perception, and cognition are dramatically influenced by
the context of events in space and time. I focus on spatial context
in vision as a paradigmatic example. Perceptually, spatial context can
induce intriguing illusions and salience phenomena, and plays a critical
role in object grouping, segmentation, and recognition. However, the
principles underlying contextual effects are not well understood.  An
appealing hypothesis suggests that neurons represent inputs in a coordinate
system that is matched to the statistical structure of images in the
natural environment. I describe a theoretical framework that is based on a
rich generative account of natural image statistics. In this framework,
Bayesian inference amounts to a generalized form of divisive normalization,
a canonical computation that has been implicated in many neural areas. In
particular, I suggest that spatial contextual influences can be switched on
or off depending on issues of grouping and segmentation in scene
statistics. I show that this approach can explain a wealth of data,
including some cortical nonlinearities, perceptual misjudgments in the tilt
illusion, and perceptual salience phenomena. The theory further makes
distinct predictions about cortical neural processing of natural images,
which we are testing through experimental collaboration.





*10:45 am*

*Jonathan Pillow*

*The University of Texas at Austin*

*"Advanced statistical methods for deciphering the neural code"*







Recent technological advances have provided increasingly detailed
measurements of the neural activity underlying various sensory, motor, and
cognitive functions. It is therefore an important challenge to develop
statistical and computational methods for making sense of complex,
high-dimensional neural signals and the information they carry.  In this
talk, I will present several projects aimed at deciphering the neural code
in different brain areas and several different levels of biophysical
detail. First, I will describe new methods for rapidly characterizing
high-dimensional response properties of neurons in visual cortex using
closed-loop "adaptive" experimental designs.  Second, I will describe
recent work on characterizing the representation of sensory-motor decisions
by neural populations in the lateral intra-parietal cortex (area LIP) in
primates.  I will discuss the implications of these findings for
understanding neural codes, the mechanisms by which they are constructed,
and the strategies by which they may be read out by other brain areas.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cs.princeton.edu/pipermail/ml-stat-talks/attachments/20140129/caaac396/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 9174 bytes
Desc: not available
URL: <http://lists.cs.princeton.edu/pipermail/ml-stat-talks/attachments/20140129/caaac396/attachment-0001.jpe>


More information about the Ml-stat-talks mailing list