This is the BBOB 2015 results page and here are the results from the workshop held at GECCO in July 2015.

Results for all algorithms

The following tables display results relative to the best result of 2009. They can be therefore more easily compared to the result tables from BBOB-2009.

  • Comparison tables per function: noiseless testbed dimensions 2-20: pdf
  • Comparison tables per function: noiseless testbed dimensions 2-20 (expensive setting): pdf
Single Function Tables

The following tables display results relative to the best result of 2009. They can be therefore more easily compared to the result tables from BBOB-2009. Note that there has been no submission for the noisy testbed at GECCO 2015 and only three of the algorithms have been benchmarked on the optional 40-dimensional problems.

Standard Setting

  • Comparison tables per function: noiseless testbed 2-D: pdf
  • Comparison tables per function: noiseless testbed 3-D: pdf
  • Comparison tables per function: noiseless testbed 5-D: pdf
  • Comparison tables per function: noiseless testbed 10-D: pdf
  • Comparison tables per function: noiseless testbed 20-D: pdf
  • Comparison tables per function: noiseless testbed 40-D: pdf

Expensive Setting

  • Comparison tables per function: noiseless testbed 2-D (expensive setting): pdf
  • Comparison tables per function: noiseless testbed 3-D (expensive setting): pdf
  • Comparison tables per function: noiseless testbed 5-D (expensive setting): pdf
  • Comparison tables per function: noiseless testbed 10-D (expensive setting): pdf
  • Comparison tables per function: noiseless testbed 20-D (expensive setting): pdf
  • Comparison tables per function: noiseless testbed 40-D (expensive setting): pdf
Available Source Codes

CMA-ES source code for running CMA-CSA, CMA-MSR, and CMA-TPA

Data
Resources: Accepted papers
  • Asma Atamna: Benchmarking IPOP-CMA-ES-TPA and IPOP-CMA-ES-MSR on the BBOB Noiseless Testbed
  • Lukáš Bajer, Zbynek Pitra, Martin Holena: Benchmarking Gaussian Processes and Random Forests Surrogate Models on the BBOB Noiseless Testbed
  • Dimo Brockhoff, Bernd Bischl, Tobias Wagner: The Impact of Initial Designs on the Performance of MATSuMoTo on the Noiseless BBOB-2015 Testbed: A Preliminary Study
  • Petr Pošík, Petr Baudiš: Dimension Selection in Axis-Parallel Brent-STEP Method for Black-Box Optimization of Separable Continuous Functions
bbob-2015-results.txt · Last modified: 2017/05/15 11:56 by brockho
CC Attribution-Noncommercial-Share Alike 3.0 Unported
Valid CSS Driven by DokuWiki Recent changes RSS feed Valid XHTML 1.0