University of Arizona

4/17/1997

3:45pm-4:45pm

There has been a recent surge in interest in combining multiple estimators, both from the perspective of neural networks and statistics as well as cognitive science, where the different estimators may represent different cues about an underlying value such as the depth of a visual surface. In this talk I will focus on the problem of statistically correct inference in networks whose basic representations are population codes.

Population codes are ubiquitous in the brain, and involve the simultaneous activity of many units coding for some low dimensional quantity. A classic example are place cells in the rat hippocampus: several of these cells fire when the animal is at a particular place in an environment, so the underlying quantity has two dimensions of spatial location.

I will show how to interpret the population activity as encoding whole probability distributions over the underlying variable rather then just single values, and propose a method of inductively learning mappings between population codes that are computationally tractable and yet offer good approximations to statistically optimal inference. I will also present some simple applications of the method to prove its competence. This is joint work with Peter Dayan.

*Refreshments will be served immediately before the talk at 3:30pm.Hosted by Michael Mozer.*

Department of Computer Science

University of Colorado Boulder

Boulder, CO 80309-0430 USA

webmaster@cs.colorado.edu

University of Colorado Boulder

Boulder, CO 80309-0430 USA

webmaster@cs.colorado.edu