News for hackers. Stuff that matters.
Re:A New Coefficient of Correlation by hinoki on Saturday December 25, 2021 @07:43PM (#29688525)
The summary says that it works if it is a measurable function [0], which is structure preserving. So sha256 would scramble the input too much for it to be detected here.

[0] https://en.m.wikipedia.org/wiki/Measurable_function

Re:A New Coefficient of Correlation by kkylin on Saturday December 25, 2021 @08:48PM (#29688941)
I've not yet read the article, just the abstract. But the abstract is pretty precise, and being "measurable" is a very weak constraint. For (computationally) complex functions like a hash, my first guess is the "escape clause" is in the number of samples needed for the statistic to converge to its expected value.
Re:A New Coefficient of Correlation by nnm on Sunday December 26, 2021 @12:31AM (#29690000)
Any function that can be implemented in a computer is a Lebesgue measurable function.
Re:A New Coefficient of Correlation by bollu on Sunday December 26, 2021 @03:23AM (#29690691)
I don't doubt it, but I don't immediately see the proof either. What's the key idea?
Re:A New Coefficient of Correlation by contravariant on Sunday December 26, 2021 @09:57AM (#29692815)
Most spaces that computers deal with are basically discrete.

Technically this may not always be the case but it's very hard to construct a convincing counter example.