All Mutation Rates $c/n$ for the $(1+1)$ Evolutionary Algorithm

TL;DR

This paper demonstrates that for any mutation rate proportional to $c/n$ with $c \\geq 1$, there exists a fitness function where this rate is nearly optimal for the $(1+1)$ EA, showing the density of optimal rates.

Contribution

The paper introduces a fitness function called HillPathJump to prove that all mutation rates $c/n$ with $c \\geq 1$ can be nearly optimal, establishing their density in the interval.

Findings

Optimal mutation rates $c/n$ are dense for $c \\geq 1$.

For each $c \\geq 1$, a corresponding fitness function exists.

The HillPathJump function demonstrates this density property.

Abstract

For every real number $c \geq 1$ and for all $\varepsilon > 0$, there is a fitness function $f : \{0,1\}^n \to \mathbb{R}$ for which the optimal mutation rate for the $(1+1)$ evolutionary algorithm on $f$, denoted $p_n$, satisfies $p_n \approx c/n$ in that $|np_n - c| < \varepsilon$. In other words, the set of all $c \geq 1$ for which the mutation rate $c/n$ is optimal for the $(1+1)$ EA is dense in the interval $[1, \infty)$. To show this, a fitness function is introduced which is called HillPathJump.