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

Probabilistic incremental program evolution

Salustowicz1, Schmidhuber

  • 1rafal@idsia.ch

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

[Prof. Wilhelm Gröschel, DDS, MD, on his 60th birthday].

Zahnarztliche Welt, zahnarztliche Rundschau, ZWR, vereinigt mit Zahnarztliche Reform·1967
Same author

[In memoriam Prof. Dr. Ludwig Schubert].

Deutsche zahnarztliche Zeitschrift·1966
Same author

[90-year-old Professor Karl Zilkens].

Zahnarztliche Mitteilungen·1966
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

Probabilistic Incremental Program Evolution (PIPE) is a novel automatic program synthesis technique. It efficiently generates better programs by adaptively refining probability distributions, outperforming genetic programming on benchmark tasks.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Automatic program synthesis aims to create functional programs from specifications.
  • Existing methods like genetic programming (GP) can be computationally intensive.
  • Efficiently discovering optimal programs, especially those with minimal runtime, remains a challenge.

Purpose of the Study:

  • Introduce Probabilistic Incremental Program Evolution (PIPE) as a novel technique for automatic program synthesis.
  • Demonstrate PIPE's efficiency and effectiveness compared to traditional methods like GP.
  • Apply PIPE to solve complex problems, including function regression, parity, and partially observable mazes.

Main Methods:

  • Combine probability vector coding of program instructions with population-based incremental learning.

Related Experiment Videos

  • Utilize tree-coded programs, similar to those in genetic programming.
  • Iteratively generate program populations using an adaptive probability distribution, refining it with the best program from each iteration.

    • Main Results:

    • PIPE stochastically generates progressively better programs.
    • PIPE efficiently evaluates program populations by focusing distribution refinements on the best program.
    • PIPE demonstrated superior performance compared to GP on function regression and the 6-bit parity problem.
    • PIPE successfully solved tasks in partially observable mazes, yielding programs with minimal runtime.

    Conclusions:

    • PIPE offers an efficient and effective approach to automatic program synthesis.
    • The adaptive probability distribution refinement in PIPE enables rapid discovery of high-performing programs.
    • PIPE's ability to optimize for minimal runtime makes it suitable for performance-critical applications.