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

Evolutionary induction of sparse neural trees

Zhang1, Ohm, Muhlenbein

  • 1Department of Computer Engineering, Seoul National University, Korea. btzhang@comp.snu.ac.kr

Evolutionary Computation
|July 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

Hot Bands of Water up to 6nu2-5nu2 in the 933-2500 cm-1 Region.

Journal of molecular spectroscopy·1999
Same author

Process engineering strategy for recombinant protein recovery from canola by cation exchange chromatography

Biotechnology progress·1999
Same author

Nanometer Deformation Caused by the Laplace Pressure and the Possibility of Its Effect on Surface Tension Measurements.

Journal of colloid and interface science·1999
Same author

The Effect of Membrane Charge on Gold Nanoparticle Synthesis via Surfactant Membranes.

Journal of colloid and interface science·1999
Same author

Error Analysis of Three Spherotensiometric Methods.

Journal of colloid and interface science·1999
Same author

Potential Physiological Activities of Fungi and Bacteria in Relation to Plant Litter Decomposition along a Gap Size Gradient in a Natural Subtropical Forest

Microbial ecology·1998
Same journal

Computing Optimal Populations for Binary Problems using Logic Minimization.

Evolutionary computation·2026
Same journal

Enhancing Generalization and Scalability for Multi-Objective Optimization with Population Pre-Training.

Evolutionary computation·2026
Same journal

XCS for Sequential Perceptual Aliasing in Multi-Step Decision Making.

Evolutionary computation·2026
Same journal

A dynamic multi-objective evolutionary algorithm using dual-space prediction and surrogate-based sampling.

Evolutionary computation·2026
Same journal

Adapting MOEA/D to CMA-ES for Dealing with Ill-conditioned Multiobjective Problems.

Evolutionary computation·2026
Same journal

Editorial of the Special Issue: Parallel Problem Solving from Nature PPSN 2024 Extended Versions of Best Paper Candidates.

Evolutionary computation·2026
See all related articles

This study introduces neural trees for parsimonious neural network induction. A hybrid evolutionary algorithm optimizes network structure and parameters for improved generalization in time series prediction.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Automatic induction of neural networks involves both structural and parametric learning.
  • Existing methods face challenges in efficiently learning network architectures and parameters simultaneously.

Purpose of the Study:

  • To present a novel representation scheme, neural trees, for efficient neural network induction.
  • To develop a hybrid evolutionary method for learning sparse and generalizable neural network structures.

Main Methods:

  • Utilizing a novel 'neural trees' representation for network architecture.
  • Employing a hybrid evolutionary algorithm combining genetic programming and breeder genetic algorithm.
  • Applying the minimum description length principle for unified learning framework.

Related Experiment Videos

Main Results:

  • Successfully induced higher-order neural trees with sparse structures.
  • Demonstrated good generalization performance on chaotic time series prediction tasks.
  • Efficiently learned both network architectures and parameters through genetic search.

Conclusions:

  • Neural trees offer an effective approach for parsimonious neural network induction.
  • The hybrid evolutionary method enhances generalization by promoting sparse network structures.
  • The approach is suitable for complex problems like chaotic time series prediction.