当前位置: X-MOL 学术IEEE T. Evolut. Comput. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A General Dichotomy of Evolutionary Algorithms on Monotone Functions
IEEE Transactions on Evolutionary Computation ( IF 14.3 ) Pub Date : 2020-12-01 , DOI: 10.1109/tevc.2019.2917014
Johannes Lengler

It is known that the (1 + 1)-EA with mutation rate $c/n$ optimizes every monotone function efficiently if $c < 1$ , and needs exponential time on some monotone functions (HotTopic functions) if $c\geq 2.2$ . We study the same question for a large variety of algorithms, particularly for the $(1 + \lambda)$ -EA, $(\mu + 1)$ -EA, $(\mu + 1)$ -GA, their “fast” counterparts, and for the $(1 + (\lambda,\lambda))$ -GA. We find that all considered mutation-based algorithms show a similar dichotomy for HotTopic functions, or even for all monotone functions. For the $(1 + (\lambda,\lambda))$ -GA, this dichotomy is in the parameter $c\gamma $ , which is the expected number of bit flips in an individual after mutation and crossover, neglecting selection. For the fast algorithms, the dichotomy is in $m_{2}/m_{1}$ , where $m_{1}$ and $m_{2}$ are the first and second falling moment of the number of bit flips. Surprisingly, the range of efficient parameters is not affected by either population size $\mu $ nor by the offspring population size $\lambda $ . The picture changes completely if crossover is allowed. The genetic algorithms $(\mu + 1)$ -GA and $(\mu + 1)$ -fGA are efficient for arbitrary mutations strengths if $\mu $ is large enough.

中文翻译:

单调函数进化算法的一般二分法

已知(1+1)-EA具有突变率 $c/n$ 有效地优化每个单调函数,如果 $c < 1$ ,并且在某些单调函数(HotTopic 函数)上需要指数时间,如果 $c\geq 2.2$ . 我们针对各种算法研究相同的问题,特别是对于 $(1 + \lambda)$ -EA, $(\mu + 1)$ -EA, $(\mu + 1)$ -GA,他们的“快速”同行,以及 $(1 + (\lambda,\lambda))$ -GA。我们发现所有考虑的基于变异的算法都对 HotTopic 函数甚至所有单调函数都显示出类似的二分法。为了 $(1 + (\lambda,\lambda))$ -GA,这个二分法在参数中 $c\gamma $ ,这是个体在变异和交叉后的预期位翻转次数,忽略选择。对于快速算法,二分法是 $m_{2}/m_{1}$ , 在哪里 $m_{1}$ $m_{2}$ 是位翻转次数的第一个和第二个下降时刻。令人惊讶的是,有效参数的范围不受任何种群大小的影响 $\亩 $ 也不是由后代人口规模 $\lambda $ . 如果允许交叉,画面会完全改变。遗传算法 $(\mu + 1)$ -GA 和 $(\mu + 1)$ -fGA 对于任意突变强度是有效的,如果 $\亩 $ 足够大。
更新日期:2020-12-01
down
wechat
bug