no code implementations • 18 Apr 2024 • Cella Florescu, Marc Kaufmann, Johannes Lengler, Ulysse Schaller
For the classical benchmark OneMax, the cGA has to two different modes of operation: a conservative one with small step sizes $\Theta(1/(\sqrt{n}\log n))$, which is slow but prevents genetic drift, and an aggressive one with large step sizes $\Theta(1/\log n)$, in which genetic drift leads to wrong decisions, but those are corrected efficiently.
1 code implementation • 13 Nov 2023 • Marc Kaufmann, Maxime Larcher, Johannes Lengler, Oliver Sieberling
Recently, Kaufmann, Larcher, Lengler and Zou conjectured that for the self-adjusting $(1,\lambda)$-EA, Adversarial Dynamic BinVal (ADBV) is the hardest dynamic monotone function to optimize.
1 code implementation • 14 Apr 2022 • Marc Kaufmann, Maxime Larcher, Johannes Lengler, Xun Zou
In this paper we disprove this conjecture and show that OneMax is not the easiest fitness landscape with respect to finding improving steps.
1 code implementation • 1 Apr 2022 • Marc Kaufmann, Maxime Larcher, Johannes Lengler, Xun Zou
Recently, Hevia Fajardo and Sudholt have shown that this setup with $c=1$ is efficient on \onemax for $s<1$, but inefficient if $s \ge 18$.
no code implementations • 9 Apr 2021 • Marc Kaufmann
Finally, when they have either increasing or decreasing productivity, people work less each day than previously planned.
1 code implementation • 12 Jan 2021 • Francesco Fallucchi, Marc Kaufmann
Many important economic outcomes result from cumulative effects of smaller choices, so the best outcomes require accounting for other choices at each decision point.