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

Completely derandomized self-adaptation in evolution strategies.

N Hansen1, A Ostermeier

  • 1Technische Universität Berlin, Fachgebiet für Bionik, Sekr. ACK 1, Ackerstr. 71-76, 13355 Berlin, Germany. hansen@bionik.tu-berlin.de

Evolutionary Computation
|May 31, 2001
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

SMAS: Structural MRI-based AD Score using Bayesian supervised VAE.

Computers in biology and medicine·2025
Same author

Gaussian Process-based prediction of memory performance and biomarker status in ageing and Alzheimer's disease-A systematic model evaluation.

Medical image analysis·2023
Same author

Molecular-Weight Growth in Ozone-Initiated Low-Temperature Oxidation of Methyl Crotonate.

The journal of physical chemistry. A·2020
Same author

Process and Outcome Evaluation of a Mindfulness-Based Cognitive Therapy Intervention for Cisgender and Transgender African American Women Living with HIV/AIDS.

AIDS and behavior·2020
Same author

Diagonal Acceleration for Covariance Matrix Adaptation Evolution Strategies.

Evolutionary computation·2019
Same author

Detection of Aliphatically Bridged Multi-Core Polycyclic Aromatic Hydrocarbons in Sooting Flames with Atmospheric-Sampling High-Resolution Tandem Mass Spectrometry.

The journal of physical chemistry. A·2018
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 derandomization and cumulation for self-adapting mutation distributions in evolution strategies. Covariance Matrix Adaptation (CMA) significantly boosts performance on challenging, non-separable functions.

Area of Science:

  • Optimization Algorithms
  • Evolutionary Computation
  • Machine Learning

Background:

  • Mutation distribution self-adaptation is crucial for efficient evolutionary algorithms.
  • Existing methods like mutative strategy parameter control have limitations.
  • Arbitrary normal mutation distributions are equivalent to general linear problem encoding.

Purpose of the Study:

  • To develop methods for self-adaptation of mutation distributions.
  • To address shortcomings in current strategy parameter control.
  • To introduce a robust self-adaptation scheme for arbitrary normal mutation distributions.

Main Methods:

  • Introduced concepts of derandomization and cumulation for self-adaptation.
  • Developed a completely derandomized self-adaptation scheme: Covariance Matrix Adaptation (CMA).

Related Experiment Videos

  • Enhanced CMA with cumulation, utilizing evolution paths instead of single steps.

    • Main Results:

    • CMA effectively adapts arbitrary normal mutation distributions.
    • Cumulation further improves CMA performance.
    • Significant speed-up factors (orders of magnitude) observed on badly scaled, non-separable functions compared to standard evolution strategies.

    Conclusions:

    • Covariance Matrix Adaptation (CMA) with cumulation offers superior performance on complex optimization problems.
    • The proposed methods provide a powerful tool for self-adaptation in evolution strategies.
    • Performance gains are most pronounced on non-separable and mis-scaled functions.