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

Centroid neural network for unsupervised competitive learning.

D C Park1

  • 1Intelligent Computing Research Lab., School of Electrical and Information Control Engineering, Myong Ji University, Yong In, Kuung Ki-do 449-728, Korea. parkd@wh.myongji.ac.kr

IEEE Transactions on Neural Networks
|February 6, 2008
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

Ectopic pregnancy occurring in the remnant tube of a previous adnexectomy: a case report.

Clinical and experimental obstetrics & gynecology·2018
Same author

Bone mineral density among Korean females aged 20-50 years: influence of age at menarche (The Korea National Health and Nutrition Examination Survey 2008-2011).

Osteoporosis international : a journal established as result of cooperation between the European Foundation for Osteoporosis and the National Osteoporosis Foundation of the USA·2017
Same author

Effects of β-Glucan on the Release of Nitric Oxide by Macrophages Stimulated with Lipopolysaccharide.

Asian-Australasian journal of animal sciences·2016
Same author

Decreased Bcl-6 and increased Blimp-1 in the peritoneal cavity of patients with endometriosis.

Clinical and experimental obstetrics & gynecology·2015
Same author

Paraneoplastic arthritis mimicking rheumatoid arthritis in cervical cancer.

European journal of gynaecological oncology·2015
Same author

Attitudes and practices of Korean gynecologists towards hormone replacement therapy in endometrial cancer survivors.

European journal of gynaecological oncology·2014
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

A new centroid neural network (CNN) offers faster, stable clustering and image compression. This unsupervised learning algorithm improves upon traditional methods like SOM and DCL without needing predefined parameters.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Unsupervised learning algorithms are crucial for data analysis and pattern recognition.
  • Traditional algorithms like K-means, Kohonen's Self-Organizing Map (SOM), and Kosko's Differential Competitive Learning (DCL) have limitations in convergence speed and stability.
  • Parameter tuning, such as learning coefficients and iteration counts, often impacts the performance of these algorithms.

Purpose of the Study:

  • To introduce a novel unsupervised competitive learning algorithm, the Centroid Neural Network (CNN).
  • To demonstrate the algorithmic relationships between CNN, SOM, and DCL.
  • To evaluate the performance of CNN in terms of convergence speed, clustering quality, and stability compared to conventional methods.

Main Methods:

Related Experiment Videos

  • Development of the Centroid Neural Network (CNN) algorithm, an unsupervised competitive learning approach.
  • Estimation of cluster centroids directly from training data.
  • Comparative analysis of CNN against SOM and DCL using clustering and image compression problems.

Main Results:

  • CNN demonstrates significantly faster convergence compared to conventional algorithms.
  • The algorithm achieves compatible clustering quality with improved stability.
  • CNN eliminates the need for a predetermined learning coefficient schedule or a fixed number of iterations, reducing sensitivity to initial conditions.

Conclusions:

  • The Centroid Neural Network (CNN) provides an efficient and stable unsupervised learning solution.
  • CNN offers a robust alternative to existing competitive learning algorithms, particularly in applications like clustering and image compression.
  • The proposed algorithm's independence from parameter scheduling enhances its practical applicability and reliability.