Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

A constructive algorithm for binary neural networks: the oil-spot algorithm.

F M Frattale Mascioli1, G Martinelli

  • 1INFOCOM Dept., Rome Univ.

IEEE Transactions on Neural Networks
|January 1, 1995
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Filgrastim XM02 (Tevagrastim®) after autologous stem cell transplantation compared to lenograstim: favourable cost-efficacy analysis.

Ecancermedicalscience·2013
Same author

Experimental demonstration of modulation instability in an optical fiber with a periodic dispersion landscape.

Optics letters·2012
Same author

Prediction of outcomes in patients with Ph+ chronic myeloid leukemia in chronic phase treated with nilotinib after imatinib resistance/intolerance.

Leukemia·2012
Same author

BCR-ABL-specific cytotoxic T cells in the bone marrow of patients with Ph(+) acute lymphoblastic leukemia during second-generation tyrosine-kinase inhibitor therapy.

Blood cancer journal·2012
Same author

Application of the whole-transcriptome shotgun sequencing approach to the study of Philadelphia-positive acute lymphoblastic leukemia.

Blood cancer journal·2012
Same author

Nilotinib in imatinib-resistant or imatinib-intolerant patients with chronic myeloid leukemia in chronic phase: 48-month follow-up results of a phase II study.

Leukemia·2012
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

This study introduces a new constructive training algorithm for supervised neural networks using a topological approach. The method dynamically builds networks, optimizing neuron count for specific data representations.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Supervised neural networks are widely used for complex pattern recognition tasks.
  • Efficient training algorithms are crucial for optimizing network performance and resource utilization.
  • Topological approaches offer novel perspectives for network construction.

Purpose of the Study:

  • To present a novel constructive training algorithm for supervised neural networks.
  • To leverage a topological approach for dynamic network construction.
  • To optimize neural networks in terms of neuron count.

Main Methods:

  • The algorithm employs a topological approach, mapping input space onto a binary hypercube.
  • It dynamically constructs a two-layer neural network by incrementally incorporating binary examples.

Related Experiment Videos

  • Real-valued data is handled through a real-to-binary codification scheme.

    • Main Results:

    • Simulations demonstrate the algorithm's effectiveness in constructing neural networks.
    • The constructed networks are optimized for neuron count when target functions have efficient halfspace union representations.
    • The approach provides a convenient method for treating real-valued data.

    Conclusions:

    • The proposed constructive training algorithm offers an efficient method for building supervised neural networks.
    • The topological approach and dynamic construction lead to optimized network architectures.
    • This method is particularly advantageous for problems with specific data structures and target functions.