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

Constructive algorithms for structure learning in feedforward neural networks for regression problems.

T Y Kwok1, D Y Yeung

  • 1Dept. of Comput. Sci., Hong Kong Univ. of Sci. and Technol., Kowloon.

IEEE Transactions on Neural Networks
|January 1, 1997
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

Cost minimization analysis of capecitabine versus 5-fluorouracil-based treatment for gastric cancer patients in Hong Kong.

Journal of medical economics·2017
Same author

A 12-year review on the aetiology and surgical outcomes of paediatric rhegmatogenous retinal detachments in Hong Kong.

Eye (London, England)·2015
Same author

The preparedness of Hong Kong medical students towards advance directives and end-of-life issues.

East Asian archives of psychiatry : official journal of the Hong Kong College of Psychiatrists = Dong Ya jing shen ke xue zhi : Xianggang jing shen ke yi xue yuan qi kan·2012
Same author

Gender differences in social network characteristics and psychological well-being among Hong Kong Chinese: the role of future time perspective and adherence to Renqing.

Aging & mental health·2006
Same author

Impact of job characteristics on psychological health of Chinese single working women.

Women & health·2001
Same author

Objective functions for training new hidden units in constructive neural networks.

IEEE transactions on neural networks·1997
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 paper surveys constructive algorithms for feedforward neural network structure learning in regression. It details incremental network growth methods and categorizes algorithms by search strategy, training, and architecture.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Feedforward neural networks are widely used for regression tasks.
  • Determining optimal network structure is a significant challenge in machine learning.
  • Existing methods often require manual configuration or extensive hyperparameter tuning.

Purpose of the Study:

  • To provide a comprehensive review of constructive algorithms for neural network structure learning.
  • To categorize and analyze different approaches to incremental network growth.
  • To highlight key issues and considerations in applying constructive algorithms to regression problems.

Main Methods:

  • Systematic literature review of constructive algorithms.
  • Formulation of network construction as a state-space search problem.

Related Experiment Videos

  • Development of a taxonomy based on state transition mapping, training algorithms, and network architectures.

    • Main Results:

    • Identification and description of various constructive algorithms.
    • Analysis of general issues in constructive algorithms, particularly search strategies.
    • Presentation of a taxonomy classifying algorithms based on key differentiating features.

    Conclusions:

    • Constructive algorithms offer an automated approach to neural network design for regression.
    • The choice of search strategy, training algorithm, and architecture significantly impacts algorithm performance.
    • Further research can benefit from this taxonomy for developing more efficient and effective structure learning methods.