CellGenetic.com
A.I. Research in Hybrid Self-Adaptive Algorithms in Dynamic Environments (0.8.2.1 Beta)
Skip Navigation Links
Evolutionary Algorithms
Skip Navigation Links

Coming Soon

Evolutionary Algorithms, as the name implies, employ evolutionary methods of reproduction through genetic crossover and mutation to procure solutions to complex problems.

The number of potential solutions in a given search landscape increases exponentially with the number of parameters and number of parameter values. 

It is very difficult, if not entirely impossible, for a person to fine tune optimum parameter configurations.  Evolutionary methods combine dna profiles of the fittest candidates to generate new candidates for hypothesis testing. 

These algorithms are still sensitive to both the initial random generation (generation zero), and to the set of parameters selectable from within the dna sequences. 

Hybrid algorithms will minimise the initial selection set, and enable the injection of novel dna into the existing profiles, that is, dna which did not exist at the time of initial solution definition. 

The hybrid cellular automata governing algorithm detects preferential selection within the solution landscape and can choose to inject new range boundaries, to split existing values, or coalesce existing values as part of its parameter control cycle. 

 

Copyright Martin Kelly 2014
You are not currently logged in
Click here for the Login Page