Effective matrix adaptation strategy for noisy derivative-free optimization

  • Morteza Kimiaei
  • Arnold Neumaier
    • Categories Global Optimization Applications, Nonlinear Optimization, Optimization Software Benchmark Tags Short URL: https://optimization-online.org/?p=22367

    In this paper, we introduce a new effective matrix adaptation evolution
    strategy (MADFO) for noisy derivative-free optimization problems. Like
    every MAES solver, MADFO consists of three phases: mutation, selection
    and recombination. MADFO improves the mutation phase by generating
    good step sizes, neither too small nor too large, that increase the
    probability of selecting mutation points with small inexact function
    values in the selection phase. In the recombination phase, recombination
    points with low inexact function values (best points) are found
    by a new randomized non-monotone line search method. If no new
    best point is found, heuristic points may be accepted as new best
    points. We compare MADFO with state-of-the-art DFO solvers on noisy
    test problems obtained by adding various kinds and levels of noise
    to all unconstrained CUTEst test problems with dimensions n ≤ 20,
    and find  that MADFO has the highest number of solved problems.

    Article

    Download

    View Effective matrix adaptation strategy for noisy derivative-free optimization