=Paper=
{{Paper
|id=Vol-2962/invited03
|storemode=property
|title=Black-Box Min-Max Optimization using CMA-ES
|pdfUrl=https://ceur-ws.org/Vol-2962/invited03.pdf
|volume=Vol-2962
|authors=Youhei Akimoto
|dblpUrl=https://dblp.org/rec/conf/itat/Akimoto21
}}
==Black-Box Min-Max Optimization using CMA-ES==
<pdf width="1500px">https://ceur-ws.org/Vol-2962/invited03.pdf</pdf>
<pre>
                             Black-Box Min-Max Optimization using CMA-ES

                                                           Yohei Akimoto
                                                           Youhei Akimoto


                                                    University of Tsukuba, Japan



Abstract: Simulation-based optimization, where the objective function value f(x) is computed through a numerical simulation given
design variable x, appears widely in engineering problems. However, it is often observed that a solution obtained by simulation-
based optimization do not perform as well on the real environment as on the simulation, due to the difference between the real
environment and the simulation. This is one of the factors that impair the reliability of the optimization result in the simulation. The
problem of finding a solution robust to such error and uncertainty is often formulated as a minimax optimization, where we try to
locate the best solution under the worst possible situation, to guarantee the worst case performance. In this talk, we present
CMA-ES, which is recognized as one of the most powerful solvers for simulation-based optimization and has been applied to many
real-world applications. Then, we present our recent developments in black-box minimax optimization using CMA-ES.




____________________
Copyright ©2021 for this paper by its authors. Use permitted under Creative Commons
License Attribution 4.0 International (CC BY 4.0).

</pre>