please explane me genethic algorithms .?

Question

such a machine learning algorithm it is.

Answer 1

Genetic Algorithms are part of evolutionary computing and contains many subsets like Genetic Programming and other approaches. Some main concepts that doesn't change are
1) representation : you must be able to represent evert possible solution in your solution space of your problem in a suitable format like a string of genomes
2)genetic operators : you must define basic genetic operations like mutation and crossover which may be applied with a predetermined possibility to every single member of population at every epoch of evolution.
3)You must have a fitness function which gives a score to each off-spring
also see roulette wheel selection, elitism, etc...

Genetic Algorithms are not deterministic and are not guaranteed to give the optimal solution but they work like a charm all the times and evolutionary computing have generated many human competitive solutions to hardest problems on the field.

Loren Soth

Answer 2

Basically, if given a problem, come up with a ton of little solutions to the problem. Now randomly combine these little solutions into plans for how to solve a problem. You can think of these plans as DNA. Now run your fitness algorithm on the DNA to see how it performs. When that's done, you must decide what plans to keep. The cycle starts over from here, you use pieces of the solutions that worked to make new plans and re-score them. Occasionally you mutate your solutions by altering their results. Eventually you should get good answers to your problem, probably solutions you didn't think of. These algorithms aren't absolutely good answers, it's sort of playing God, but we are just human so we can't completely do that...

Answer 3

GAs take more than a quick synopsis to fully explain. In short it the evolution of a population of solutions to a given problem via crossover and mutation. This I'm sure doesn't help you, so instead try searching the net. You can spend all your life studying a subject as diverse as GAs and still not know the half of it, so to expect to have it all explained to you here is a little silly :P Good luck!