Edinburgh Research Explorer

A Hierarchical Generative Model of Recurrent Object-Based Attention in the Visual Cortex

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Original languageEnglish
Title of host publicationARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2011, PT I
EditorsT Honkela, W Duch, M Girolami, S Kaski
Place of PublicationBERLIN
PublisherSpringer-Verlag Berlin Heidelberg
Pages18-25
Number of pages8
ISBN (Print)978-3-642-21734-0
Publication statusPublished - 2011
Event21st International Conference on Artificial Neural Networks, ICANN 2011 - Espoo, Finland
Duration: 14 Jun 201117 Jun 2011

Publication series

NameLecture Notes in Computer Science
PublisherSPRINGER-VERLAG BERLIN
Volume6791
ISSN (Print)0302-9743

Conference

Conference21st International Conference on Artificial Neural Networks, ICANN 2011
CountryFinland
Period14/06/1117/06/11

Abstract

In line with recent work exploring Deep Boltzmann Machines (DBMs) as models of cortical processing, we demonstrate the potential of DBMs as models of object-based attention, combining generative principles with attentional ones. We show: (1) How inference in DBMs can be related qualitatively to theories of attentional recurrent processing in the visual cortex; (2) that deepness and topographic receptive fields are important for realizing the attentional state; (3) how more explicit attentional suppressive mechanisms can be implemented, depending crucially on sparse representations being formed during learning.

    Research areas

  • RECOGNITION

Event

ID: 20027427