• Home
  • Search
  • Translation-Invariant Neural Responses as Variational Messages in a Bayesian Network Model
  • https://doi.org/10.1007/978-3-319-68600-4_20Copy DOI Icon

Translation-Invariant Neural Responses as Variational Messages in a Bayesian Network Model

  • Jan 1, 2017
  • Takashi Sano +1 more
Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

In this paper, we propose the following interpretation: if a Bayesian network has acquired translation invariance in input images, its feedback messages from higher layers to lower layers can be interpreted as the response of complex cells in the visual system. To examine our proposal’s validity, we trained a Bayesian network to acquire translation invariance using the standard belief propagation algorithm, and confirmed its feedback messages were translation invariant and thus they can be interpreted as the response of complex cells. Unlike previous studies, our model does not require specially prepared random variables. Furthermore, our model only uses the standard belief propagation algorithm. Therefore we believe that our model is more natural than the previous ones to integrate hierarchical Hubel-Wiesel architectures for the visual system, e.g. Hierarchical MAX models, and probabilistic graphical models.

Similar Papers
  • Book Chapter
  • Citations44

Background and perspectives of possibilistic graphical models

  • Jan 01, 1997
  • Jörg Gebhardt +1
  • Book Chapter
  • Citations2

Probabilistic Graphical Models for Medical Image Mining Challenges of New Generation

  • Jan 01, 2018
  • Sridevi Tumula +1
  • Research Article
  • Citations90

Bayesian networks of customer satisfaction survey data

  • Nov 01, 2009
  • Journal of Applied Statistics
  • Silvia Salini +1
  • Research Article
  • Citations451

Bayesian Networks in Fault Diagnosis

  • Oct 01, 2017
  • IEEE Transactions on Industrial Informatics
  • Baoping Cai +2
  • Research Article
  • Citations2

Modelling structured data with Probabilistic Graphical Models

  • Jan 01, 2016
  • EAS Publications Series
  • F Forbes
  • Research Article
  • Citations133

Non-parametric Bayesian networks: Improving theory and reviewing applications

  • Aug 14, 2015
  • Reliability Engineering & System Safety
  • Anca Hanea +2
  • Book Chapter
  • Citations205

An Introduction to Probabilistic Graphical Models

  • Jan 01, 2002
  • P Larrañaga
  • Supplementary Content

Deep Probabilistic Graphical Modeling.

  • Jul 03, 2020
  • arXiv (Cornell University)
  • Adji B Dieng
  • PDF
  • Research Article
  • Citations21

Inference attacks on genomic data based on probabilistic graphical models

  • Jul 16, 2020
  • Big Data Mining and Analytics
  • Zaobo He +1
  • Research Article
  • Citations13

Probabilistic Graphical Models for Computational Biomedicine

  • Jan 01, 2003
  • Methods of Information in Medicine
  • P Antal +3
  • Research Article
  • Citations12

PyBNesian: An extensible python package for Bayesian networks

  • Jul 08, 2022
  • Neurocomputing
  • David Atienza +2
  • PDF
  • Research Article
  • Citations5

Dynamic decorrelation as a unifying principle for explaining a broad range of brightness phenomena.

  • Apr 26, 2021
  • PLOS Computational Biology
  • Alejandro Lerer +2
  • Research Article
  • Citations19

Bayesian network structure training based on a game of learning automata

  • Jan 12, 2016
  • International Journal of Machine Learning and Cybernetics
  • S Gheisari +3
  • Book Chapter
  • Citations1

Foveated Vision for Biologically Inspired Continuous Face Authentication

  • Jan 01, 2019
  • Souad Khellat-Kihel +2
  • Conference Article
  • Citations16

Root cause analysis using artificial intelligence

  • Jan 01, 2017
  • Asha Chigurupati +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.