I am trying to understand the difference between biological and artificial evolution. If we look at it in terms of genetics, in both of them, the selection operation is a key term.
What's the difference between biological and artificial evolution?
$\begingroup$
$\endgroup$
0
Add a comment
|
2 Answers
$\begingroup$
$\endgroup$
7
Biological and artificial evolution work around pretty much the same principles.
Fitness and selection: In biology, the fittest organisms in an ecosystem are more likely to survive long enough to reproduce, passing on their genes in the process. In artificial evolution, our organisms are in fact solutions to our problem, which can be evaluated to determine how good they are (their fitness). We choose ourselves which solutions will be selected for reproduction (there are many ways to do this selection, but what is common among all of them is that the fittest solutions have a higher chance of being selected).
Crossover: In biology, an organism inherits a portion of each parent's genes, so is a sort of genetic hybrid of both parents. For artificial evolution, a new solution (a "child" solution) will inherit part of its parent's solutions (we take a partial solution from each parent, and glue those partial solutions together to construct a new solution).
Mutation: In nature mutations often occur at birth and this is why there are many different species. Harmful mutations make the individual less likely to survive long enough to pass them on to children, and in contrast helpful mutations make it more likely that the individual will survive long enough to pass them unto children. The same can be said for artificial evolution: A mutation randomly changes a small part of the solution, and if it makes that solution fitter, then that solution has a higher chance of being selected for reproduction.
-
$\begingroup$ If we consider artificial selection or "selective breeding" then it'll fall in which part of the evolution? $\endgroup$ Mar 14, 2018 at 13:59
-
$\begingroup$ @GermaVinsmoke I'm not very familiar with the fields of biology and natural history, but I would say that "artificial selection" falls under a general biological/natural definition of "fitness" (e.g., a strong deer fighting a weak deer for mating rights with a female is more likely than his opponent to win the fight and breed, as he is "fitter"). $\endgroup$ Mar 14, 2018 at 15:17
-
$\begingroup$ Great answer...I think you missed one point about death..In bio evolution we can't go back to the original but in computer we can go back to previous versions...Am I correct? $\endgroup$– user9947Mar 15, 2018 at 12:36
-
$\begingroup$ @DuttaA Quite the contrary in my opinion. In biological evolution an organism could in theory transform back to something close to a previous form, if the selection pressure pushed it that way (a changing ecosystem means a changing definition of "fitness"). In contrast, with artificial evolution we are solving a specific problem so the "ecosystem" (the problem) is static, as is its objective. This means that since our solutions are getting fitter and fitter in relation to this static objective, it is unlikely that our method will drive us towards past solutions which we have already explored. $\endgroup$ Mar 15, 2018 at 16:19
-
$\begingroup$ @PhilippeOlivier I really don't think that happens, since in genetics they are always looking for a better form, we can't go back to chimpanzees however we try, since chimpanzee is not the best form, like entropy always increases and by carnot cycle you can never reach the initial form, I think the same goes for bio evolution...though I maybe wrong $\endgroup$– user9947Mar 15, 2018 at 16:37
$\begingroup$
$\endgroup$
Phillipe's excellent answer covers the crux of the subject, so I'm just going to state the obvious: the key difference is the medium and timescale.
Biological evolution is a function of the natural world, and typically occurs over a long time span, depending on the organisms and how quickly they produce new generations. (We typically think of biological evolution as occurring over "millions of years", but it can happen much more quickly, for instance in the case of microorganisms.)
Generic algorithms utilize a computing medium, which in the current era is silicon based, and involves microprocessors and various mediums for memory (magnetic tape and more recently solid-state.)
Both natural and artificial evolution are constrained by the size of the system (a planet or ecosystem in the former case, and available memory in the latter.) However:
- Artificial evolution can occur at an artificially accelerated pace, dependent on available processing resources.
This capacity for computationally "accelerated subjective time" and accelerated evolution of algorithms is on of the bases for the theory of the "technological singularity".
It might be argued that genetic engineering allows accelerated evolution for biological species, but that would not fall under natural evolution.
