I think it sure can. Randomly modify a binary function over a million potential permutations and measure the distribution of output, subset from the distributio

I think it sure can. Randomly modify a binary function over a million potential permutations and measure the distribution of output, subset from the distribution the binary function(s) that output ahead of your threshold. Should give you a fast function that does what you want with absolutely indescribable code, which isn’t an issue if it does what you want with reasonable performance.