Revision history[back]
1. As far as I remember, Genetic Programming is used for tree like structures, but is nothing but a special case of genetic algorithms.

2.If it is not a tree like structure, it might not be solved other way but with GP 3. Here there are some variations in Wikipedia 4. GA principle is to find an optimum for a given problem. So you need : a function to optimize, a good mutation function, and a lot of luck. by principle GA do not requires tremendous amounts of information, since you already know the function to be optimized, the problem with Neural networks is that you are trying to infer that function, so you need lots of data that gives you the best possible approximation to your space.

1. 95

1. As far as I remember, Genetic Genetic Programming is used for tree like like structures, but is nothing but a a special case of genetic algorithms.algorithms.95

2.If it is not a tree like structure, it might not be solved solved other way but with GP 3. Here there are some variations in Wikipedia 4. GA principle is to find an optimum for a given problem. So you you need : a function to optimize, a a good mutation function, and a lot of of luck. by principle GA do not not requires tremendous amounts of of information, since you already know know the function to be optimized, the the problem with Neural networks is that that you are trying to infer that that function, so you need lots of data data that gives you the best possible possible approximation to your space.

1. 95

1. As far as I remember, Genetic Genetic Programming is used for tree like like structures, but is nothing but a a special case of genetic algorithms.95algorithms.

2. 2.If If it is not a tree like structure, it might not be solved solved other way but with GP 3. GP.

3. Here there are some variations in Wikipedia 4.

4. GA principle is to find an optimum for a given problem. So you you need : a function to optimize, a a good mutation function, and a lot of of luck. by principle GA do not not requires tremendous amounts of of information, since you already know know the function to be optimized, the the problem with Neural networks is that that you are trying to infer that that function, so you need lots of data data that gives you the best possible possible approximation to your space.

User submitted content is under Creative Commons: Attribution - Share Alike; Other things copyright (C) 2010, MetaOptimize LLC.