> in genetic programming the goal is to find/optimize dataset to fit given algorithm No. Possibly you're confused between GAs and GP, a common confusion. In GP

> in genetic programming the goal is to find/optimize dataset to fit given algorithm No. Possibly you're confused between GAs and GP, a common confusion. In GP, the goal is to find an algorithm - in a real programming language, not as weights - to optimise an objective function. Often that is specified by input-output pairs, similar to supervised learning.