I couldn't help but notice a resemblance between what you are describing and a lot of machine learning algorithms. It seems like the most successful ML algorit

I couldn't help but notice a resemblance between what you are describing and a lot of machine learning algorithms. It seems like the most successful ML algorithms are self-correcting ones, rather than ones that attempt to calculate the exact correct answer in one go. Coincidence?