with evolutionary algorithms, i understand the concept of recombination and mutation, and that offspring may essentially constitute or approximate - be seen as - sub-populations.
but, what if your optimization problem really constitutes, and the initial population is best comprised out, of a combination of sub-populations?
creating a number of sub-populations, and combining them.
i suppose one can see this notion of sub-populations as a characteristic one would recombine and mutate on/ according to.
Hi, if you run the algorithm using all the samples of the different sub populations you increase the size of the initial population and mutations will occur between the different samples. At the opposite you can perform different runs using a different sub population for each run. Mutations between sub populations need still to be defined in your approach.
Sub-populations (elites) could keep the some good characteristics of the populations. By combining the normal population or mutated populations can help to inherit the good characteristics (gene). Please see below for detailed discussion.
Article Evolutionary-Algorithm-Assisted Joint Channel Estimation and...
This isn't necessarily an easy question to answer without ignoring important dynamics. In a heterogeneous population which subject to some set of selection biases on both attribute retention (breeding with trait transmission) and loss, it is likely that there will be a number of quite different "locally optimal" solutions in the search space. It is easy to see that this is the case because we observe speciation ... in fact, the Galapagos finches that Darwin found are a beautiful example.
In this light, using many sub-populations is probably going to cover a broader portion of your "solution space".
You don't really indicate to what degree you have mutations in traits (as opposed to just changin a mix of static traits). In an economic context, you might have a number of governing attributes, such as how risk averse an actor is, and also more complex behaviours such as an actor's purchasing behaviour that are constituted by groups of traits. This is analogous to organisms -- some are fast or slow, and there are lots of different strategies for selecting and acquiring food.
Simulations starting with many completely random traits will probably take rather longer to "settle" -- if it does -- but it is much more likely to come up with completely novel ways of dealing with things.
sub-populations are perhaps not the right word to express what i mean. sub-groups or sub-groupings is likely a better word or description
- Badges
- Science topic
- Similar topics
- Mathematics
- Applied Mathematics