Details on the Differences Between the DTS-CMA-ES from 2017 and the Updated 2018 Version

[from the authors]

  • new selection of points to be evaluated by the original fitness using an uncertainty criterion: selecting the union of the $k$ nearest neighbors of every point for which the fitness should be predicted ($k$ is maximal such that the total number of selected points does not exceed $N_{max}$)
  • bugfix in the surrogate selection algorithm (compared to the 2017 version)
  • sligthly different parameter settings (maximum radius for selecting points to the training set = 4 times the 0.99-quantile of the length of vectors from $N(\mathbf{0},\mathbf{I})$, maximum number of points in the training set = $20 D$, probability of improvement as the surrogate selection criterion)
detailsdts.txt · Last modified: 2019/03/14 16:23 by brockho
CC Attribution-Noncommercial-Share Alike 3.0 Unported
Valid CSS Driven by DokuWiki Recent changes RSS feed Valid XHTML 1.0