Subtractive Perceptrons for Learning Images: A Preliminary Report

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

Abstract

In recent years, artificial neural networks have achieved tremendous success for many vision-based tasks. However, this success remains within the paradigm of weak AI where networks, among others, are specialized for just one given task. The path toward strong AI, or Artificial General Intelligence, remains rather obscure. One factor, however, is clear, namely that the feed-forward structure of current networks is not a realistic abstraction of the human brain. In this preliminary work, some ideas are proposed to define a subtractive Perceptron (s-Perceptron), a graph-based neural network that delivers a more compact topology to learn one specific task. In this preliminary study, we test the s-Perceptron with the MNIST dataset, a commonly used image archive for digit recognition. The proposed network achieves excellent results compared to the benchmark networks that rely on more complex topologies.

Original languageEnglish (US)
Title of host publication2019 9th International Conference on Image Processing Theory, Tools and Applications, IPTA 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728139753
DOIs
StatePublished - Nov 2019
Event9th International Conference on Image Processing Theory, Tools and Applications, IPTA 2019 - Istanbul, Turkey
Duration: Nov 6 2019Nov 9 2019

Publication series

Name2019 9th International Conference on Image Processing Theory, Tools and Applications, IPTA 2019

Conference

Conference9th International Conference on Image Processing Theory, Tools and Applications, IPTA 2019
Country/TerritoryTurkey
CityIstanbul
Period11/6/1911/9/19

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Signal Processing

Fingerprint

Dive into the research topics of 'Subtractive Perceptrons for Learning Images: A Preliminary Report'. Together they form a unique fingerprint.

Cite this