Diversity Priors for Learning Early Visual Features

Diversity Priors for Learning Early Visual Features

This paper investigates how utilizing diversity priors can discover early visual features thatresemble their biological counterparts. The study is mainly motivated by the sparsity andselectivity of activations of visual neurons in area V1. Most previous work on computationalmodeling emphasizes selec...

Full description

Bibliographic Details
Main Authors: Hanchen eXiong, Antonio J Rodriguez-Sanchez, Sandor eSzedmak, Justus ePiater
Format: Article
Language:English
Published: Frontiers Media S.A. 2015-08-01
Series:Frontiers in Computational Neuroscience
Subjects:
inhibition
Restricted Boltzmann Machine
Markov Networks
diversity prior
V1 simple cell
Online Access:http://journal.frontiersin.org/Journal/10.3389/fncom.2015.00104/full

Internet

http://journal.frontiersin.org/Journal/10.3389/fncom.2015.00104/full

Similar Items