-1
$\begingroup$

I am discovering the topic of Genetic Algorithms, I read a bit about it on wikipedia and towardsdatascience. When I checked papers some papers, I found them using the notation "$(1 + (λ, λ))$" Genetic Algorithm. What does it mean exactly?

They also use the notation "(1 + 1)" evolutionary algorithm. Any idea what these notations mean?

Example paper: https://arxiv.org/pdf/2004.08664.pdf

Thanks

Improve this question
$\endgroup$
1
  • $\begingroup$ It's in the preliminaries of the very paper you provide. Read sections 2.2 and 2.3. $\endgroup$
    – Pål GD
    Feb 13, 2022 at 14:27

1 Answer 1

Reset to default
1
$\begingroup$

The very first sentence in the paper you link to says

The $(1 + (\lambda, \lambda))$ genetic algorithm (GA), proposed in [12], is a fairly recent algorithm with very interesting properties.

We search for the given citation, and see that it says in Section 1.3 that

We use a uniform crossover that takes bits from the parent with probability $1 − c$ and from the winning offspring with probability $c$ for some not too large crossover probability $c$. The outcome of such a crossover step will be close enough to the parent to give us a good chance of keeping the positive aspects of the parent. To give newly found positive genes of the winning offspring a reasonable chance to survive, we create again $\lambda$ offspring by this crossover. We call this algorithm the $(1 + (\lambda, \lambda))$ GA.

You can do a similar lookup for the (1+1) EA algorithm.

Improve this answer
$\endgroup$
2
  • 1
    $\begingroup$ Just to add on, $(\mu + \lambda)$ indicates a parent population size of $\mu$ generating $\lambda$ offspring each generation, with the plus indicating that the next generation comes from selecting from both parents and children (i.e., elitism). A $(\mu, \lambda)$ EA is the same except only offspring are sampled to form the next generation. $\endgroup$
    – deong
    Feb 13, 2022 at 15:57
  • 1
    $\begingroup$ @deong Care to make that (perhaps expanded a little) into an answer? I'd be happy to upvote. $\endgroup$
    – Juho
    Feb 13, 2022 at 16:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.