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

Contrast enhancement for backpropagation.

T M Kwon1, H Cheng

  • 1Dept. of Electr. and Comput. Eng., Minnesota Univ., Duluth, MN.

IEEE Transactions on Neural Networks
|January 1, 1996
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

Intraspinal leakage of bone cement after vertebroplasty: a report of 3 cases.

AJNR. American journal of neuroradiology·2006
Same author

Direct evidence of host genome acquisition by the alphaherpesvirus Marek's disease virus.

Archives of virology·2005
Same author

Functional recovery of chronic complete idiopathic transverse myelitis after administration of neurotrophic factors.

Spinal cord·2005
Same author

Cost effectiveness and delivery study for future HIV vaccines.

AIDS (London, England)·2005
Same author

[Separation of alpha-amino acids and peptides by chelated metal ion affinity chromatography].

Se pu = Chinese journal of chromatography·2005
Same author

Digital redesign of observer-based weighting switch controller for cascaded analog systems with state saturation and external loads.

ISA transactions·2005
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 modified histogram equalization technique to improve backpropagation network learning efficiency. The method preprocesses data to optimize sigmoid function activation, enhancing learning speed and balance for skewed distributions.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Neural Networks

Background:

  • Backpropagation (BP) networks are fundamental in machine learning.
  • Standard BP learning efficiency can be limited by input data characteristics.
  • Data transformation aims to optimize the sigmoid function's high-slope region for effective weight updates.

Purpose of the Study:

  • To analyze the impact of data contrast on backpropagation network performance.
  • To introduce a novel data preprocessing algorithm for enhancing BP learning efficiency.
  • To address challenges posed by skewed data distributions in BP network training.

Main Methods:

  • The study proposes a data preprocessing algorithm based on modified histogram equalization.
  • This technique transforms input data into a range that maximizes sigmoid function activation.

Related Experiment Videos

  • It specifically targets heavily concentrated data regions in skewed distributions to improve data point spacing.

    • Main Results:

    • The modified histogram equalization enhances data point spacing in concentrated regions.
    • This preprocessing improves the efficiency and balance of standard BP learning.
    • The algorithm effectively handles skewed data distributions, a common challenge in machine learning.

    Conclusions:

    • The proposed data preprocessing method significantly improves backpropagation network learning.
    • Modified histogram equalization is an effective technique for optimizing data for neural networks.
    • This approach offers a practical solution for enhancing BP network performance with challenging datasets.