Linear Readout of Object Manifolds
Disordered Systems and Neural Networks
2016-08-23 v2 Statistical Mechanics
Neural and Evolutionary Computing
Neurons and Cognition
Machine Learning
Abstract
Objects are represented in sensory systems by continuous manifolds due to sensitivity of neuronal responses to changes in physical features such as location, orientation, and intensity. What makes certain sensory representations better suited for invariant decoding of objects by downstream networks? We present a theory that characterizes the ability of a linear readout network, the perceptron, to classify objects from variable neural responses. We show how the readout perceptron capacity depends on the dimensionality, size, and shape of the object manifolds in its input neural representation.
Keywords
Cite
@article{arxiv.1512.01834,
title = {Linear Readout of Object Manifolds},
author = {SueYeon Chung and Daniel D. Lee and Haim Sompolinsky},
journal= {arXiv preprint arXiv:1512.01834},
year = {2016}
}
Comments
5 pages, 3 figures, accepted in Physical Review E as Rapid Communication on 14th May. 2016