Anonymous user
Talk:Evolutionary algorithm: Difference between revisions
→Are the Python and C++ solution cheating?: the first python code does look like it, though not intentionally
(→"Official" Algorithm: why Dawkins?) |
(→Are the Python and C++ solution cheating?: the first python code does look like it, though not intentionally) |
||
Line 109:
In answer to the first questioner of this section, the target is clearly given and is a static value. This is a major departure from what happens naturally. This is a task to show evolution, as in the gradual development of an answer towards a goal and shouldn't be taken as the answer to evolution theory sceptics. There is no intended cheating in the Python solution, it just "is what it is" and was written to follow the task goals.<br>
I wrote the task and the Python solution without being being an expert in evolutionary algorithms. It may be that what the knowledgable call an evolutionary algorithm has necessary aspects that are not part of this task description - if so, then I apologise; but I have tried to make the Python solution fit the task description as given, and did do some research into the subject at the time. --[[User:Paddy3118|Paddy3118]] 21:34, 1 October 2011 (UTC)
: That's the thing, biological evolution is not "a gradual development of an answer towards a goal". It's a process of spontaneous mutation coupled with a selection mechanism, where a mutation can go any direction, but some mutations have less of a chance to survive. This is a purely statistical process: the fitness of offsprings of previous generation are distributed around the parent (inheritance + random small mutations), but survival chance of some of them are higher than others, so only the more optimal ones will remain when it's time to create next generation, at which time the average fitness of the current generation is ''likely'' higher than the parent generation. As a ''side effect'', the average fitness of each generation exhibits a tendency towards a local maximum even though mutations are not directed.
: Once the species approaches a local maximum, random mutations still happen at the same pace, but deviating too far from the optimal configuation will lower survival chance, so the average fitness will stay there from then on. Individual specimen ''do not'' consciously regulate the mutation rate: there's no need. There's no goal in mind, it just so happens that as a collective, a species will eventually reach some average state that looks like an locally optimal solution and stay there (we call those that didn't make it "extinct").
: As a side note, I've been careful in using the word "locally". Evidently most species did not exhibit a tendency to evolve into something that's capable of inventing nuclear bombs and wielding world wars, which is clearly a globally optimal configuation. This is not relevant to current subject, however. --[[User:Ledrug|Ledrug]] 02:11, 2 October 2011 (UTC)
== "Official" Algorithm ==
| |||