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

Neuroevolution and complexifying genetic architectures for memory and control tasks.

Benjamin Inden1

  • 1Max Planck Institute for Mathematics in the Sciences, Inselstr. 22, 04103, Leipzig, Germany. inden@mis.mpg.de

Theory in Biosciences = Theorie in Den Biowissenschaften
|April 17, 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

Machine learning of symbolic compositional rules with genetic programming: dissonance treatment in Palestrina.

PeerJ. Computer science·2023
Same author

Evolving neural networks to follow trajectories of arbitrary complexity.

Neural networks : the official journal of the International Neural Network Society·2019
Same author

Evolving neural fields for problems with large input and output spaces.

Neural networks : the official journal of the International Neural Network Society·2012
Same journal

From episodes to populations: evolutionary explanation requires a constructive epistemology.

Theory in biosciences = Theorie in den Biowissenschaften·2026
Same journal

Cortical neuron classes and recursive curvature collapse: a neurobiological model of conscious dynamics.

Theory in biosciences = Theorie in den Biowissenschaften·2026
Same journal

On model of weight gain of farm animals.

Theory in biosciences = Theorie in den Biowissenschaften·2026
Same journal

An investigative network analysis mapping global cancer epidemiology.

Theory in biosciences = Theorie in den Biowissenschaften·2026
Same journal

The challenge of distinguishing living from non-living entities.

Theory in biosciences = Theorie in den Biowissenschaften·2026
Same journal

Red fescue (Festuca rubra L.) variety recognition using subset division and neural networks.

Theory in biosciences = Theorie in den Biowissenschaften·2026
See all related articles

This study introduces neuroevolution with ontogeny (NEON), an artificial evolutionary system for neural networks. NEON uses indirect encoding, offering a flexible approach to evolving complex artificial intelligence phenotypes.

Area of Science:

  • Artificial Intelligence
  • Computational Neuroscience
  • Evolutionary Computation

Background:

  • Gene interpretation in artificial evolutionary systems influences phenotype development.
  • Direct encoding methods represent genes as neurons and synapses.
  • Artificial ontogeny methods use genomes as developmental recipes.

Purpose of the Study:

  • Introduce neuroevolution with ontogeny (NEON), an indirect encoding system.
  • Demonstrate NEON's ability to emulate direct encoding methods like NEAT.
  • Evaluate NEON's performance on control and memory benchmark tasks.

Main Methods:

  • Developed NEON, a neuroevolution system employing artificial ontogeny.
  • Compared NEON's indirect encoding with direct encoding methods.

Related Experiment Videos

  • Analyzed the incremental introduction of ontogeny characteristics during evolutionary search.

    • Main Results:

    • NEON successfully emulates the neuroevolution of augmenting topologies (NEAT) method.
    • Achieved strong performance on challenging control and memory benchmark tasks.
    • Demonstrated the utility of indirect encoding through artificial ontogeny.

    Conclusions:

    • NEON provides a flexible and effective alternative to direct encoding in neuroevolution.
    • Artificial ontogeny characteristics can be incrementally integrated into evolutionary search.
    • This approach advances the development of sophisticated artificial neural networks.