

Prof. Dr. rer. nat. habil. HansGeorg Beyer

Downloads
1. Evolution Strategy Examples (for teaching only)
For the underlying algorithmic ideas refer to Scholarpedia's Evolution
Strategies.
Mathematica Sources
Matlab/Octave Sources
2. Matrix Adaptation Evolution Strategies (MAES)
Fast MAES with O(N^2) algorithmic complexity
A tarball containing Matlab/Octave code that avoids matrixmatrix
multiplication (see Slide 32 in my tutorial Design
Principles for Matrix Adaptation Evolution Strategies).
Experiments can be started in Experiment.m
Meta MAES aka BiPopulation MAES (BiPopMAES)
In order to optimize highly multimodal optimization problems, the
population size must be learned onthefly. This can be done by
running two MAES with restarts (for details see Simplify
Your Covariance Matrix Adaptation Evolution Strategy). A
tarball containing Matlab/Octave code is provided below.
Experiments can be started in BiPop_MAES_test.m
Limited Memory MAES (LMMAES)
The matrixvector operations are the algorithmic bottleneck in the
fast MAES. If one wants to reduce the algorithmic complexity below
O(N^2) one must approximate these operations. It turns out that one
can reduce this complexity to O(N log N) by introducing a set of
path cumulation vectors at different time scales to approximate the
matrix vector operations for generation of mutations. The resulting
LMMAES has been published in Large
Scale BlackBox Optimization by LimitedMemory Matrix Adaptation).
This algorithm is able to handle 10000 variables (if efficiently
coded in C or Fortran). A tarball containing Matlab/Octave code
(suited for somewhat smaller problem sizes) is provided below.
Experiments can be started in Test_LM_MA_ES.m
3. Lecture Slides (password protected)
TLV FZ PPE
last change: 14.07.2020
