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

An approach to biological computation: unicellular core-memory creatures evolved using genetic algorithms.

H Suzuki1

  • 1ATR Human Information Processing Research Laboratories, 2-2 Hikaridai Seika-cho, Soraku-gun, Kyoto 619-0288 Japan. hsuzuki@hip. atr.co.jp.

Artificial Life
|June 1, 2000
PubMed
Summary

This study introduces a novel machine learning system using genetic programming and core memories. The system successfully evolved programs to solve a machine learning problem by mimicking biological processes.

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

Purification and characterization of DNA topoisomerase IV in Escherichia coli.

The Journal of biological chemistry·1992
Same author

Identification of inducible calmodulin-dependent nitric oxide synthase in the liver of rats.

The Journal of biological chemistry·1992
Same author

Inhibition of cell attachment, invasion and metastasis of human carcinoma cells by anti-integrin beta 1 subunit antibody.

Japanese journal of cancer research : Gann·1992
Same author

Enzymatic inactivation of tachykinin neurotransmitters in the isolated spinal cord of the newborn rat.

Neuroscience research·1992
Same author

Alpha-fetoprotein-producing immature mediastinal teratoma showing rapid and massive recurrent growth in an adult.

Acta pathologica japonica·1992
Same author

Ki-67 antibody labeling index in colorectal carcinoma.

Journal of clinical gastroenterology·1992

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Bio-inspired Computing

Background:

  • Traditional programming paradigms face limitations in complex problem-solving.
  • Biological systems offer sophisticated models for adaptation and evolution.
  • Core memory technology provides a unique substrate for computational processes.

Purpose of the Study:

  • To propose and simulate a novel machine language genetic programming system.
  • To model computational processes using a biochemical reaction space analogy.
  • To investigate the efficacy of genetic algorithms in evolving machine programs.

Main Methods:

  • A one-dimensional core memory system was designed, incorporating four data word types: Membrane, Pure data, Operator, and Instruction.

Related Experiment Videos

  • Programs were represented as sequences of Instructions, processed by Operators within the core.
  • Hierarchical partitioning and tree data-flow structures were implemented using Membrane and channel Operators.
  • Genetic algorithms were employed to evolve programs based on fitness calculated from excreted Pure data.

    • Main Results:

    • The system successfully simulated the proposed genetic programming approach.
    • Programs (creatures) were evolved to solve a simple machine learning problem.
    • The fitness of programs was effectively determined by the Pure data they produced.
    • Successful breeding of programs capable of outputting predefined answers was achieved.

    Conclusions:

    • The novel core memory-based genetic programming system demonstrates potential for evolving machine intelligence.
    • The bio-inspired approach offers a new perspective on computational system design.
    • Further research can extend this system for more complex machine learning tasks.