Improving Nevergrad’s Algorithm Selection Wizard NGOpt Through Automated Algorithm Configuration

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

  • External authors:
  • Risto Trajanov
  • Ana Nikolikj
  • Gjorgjina Cenikj
  • Fabien Teytaud
  • Mathurin Videau
  • Olivier Teytaud
  • Tome Eftimov
  • Carola Doerr

Abstract

Algorithm selection wizards are effective and versatile tools that automatically select an optimization algorithm given high-level information about the problem and available computational resources, such as number and type of decision variables, maximal number of evaluations, possibility to parallelize evaluations, etc. State-of-the-art algorithm selection wizards are complex and difficult to improve. We propose in this work the use of automated configuration methods for improving their performance by finding better configurations of the algorithms that compose them. In particular, we use elitist iterated racing (irace) to find CMA configurations for specific artificial benchmarks that replace the hand-crafted CMA configurations currently used in the NGOpt wizard provided by the Nevergrad platform. We discuss in detail the setup of irace for the purpose of generating configurations that work well over the diverse set of problem instances within each benchmark. Our approach improves the performance of the NGOpt wizard, even on benchmark suites that were not part of the tuning by irace.

Bibliographical metadata

Original languageEnglish
Title of host publicationPPSN 2022: Parallel Problem Solving from Nature – PPSN XVII
Subtitle of host publication17th International Conference, PPSN 2022, Dortmund, Germany, September 10–14, 2022, Proceedings, Part I
EditorsGünter Rudolph, Anna V. Kononova, Hernán Aguirre, Pascal Kerschke, Gabriela Ochoa, Tea Tušar
Place of PublicationCham, Switzerland
PublisherSpringer Nature
Chapter2
Pages18-31
Number of pages13
VolumePPSN 2022
ISBN (Electronic) 9783031147142
ISBN (Print)9783031147135
DOIs
Publication statusPublished - 14 Aug 2022

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Nature
Volume13398
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349