Individual particle interactions are not chaotic. Simulating them one timestep at a time would take linear time in the number of particles.
They're only chaotic if you treat them in aggregate, which a superintelligence wouldn't do. It would be less lossy to get all the positions of the particles and figure out exactly what each one would do.
Something has to compute the universe, since it is currently running...
Not necessarily. There have been a number of papers trying to emphasize the generalization power of a network rather than "it got a SOTA number", and so leave off fine tuning results. Because that would distract from the point of the paper.
Pretty much all of the relationship to practical stuff in papers like this related to the RH are "it gives us information about the prime numbers, and those are used in cryptography".
It's really just about getting people to perk their ears up rather than true implications about crypto.
Agreed. Plus, its really annoying to open it on mobile and my phone starts downloading the pdf file immediately, which I don't want to have to manually delete in the future.
A related idea that I had while reading the article is this: it's easy to get computers and machines to talk to each other both unnaturally (with our own protocols) and naturally (using a randomly initialized and then learned/iterated upon version - see https://blog.openai.com/learning-to-communicate/ or Facebook's recent experiment. It's also easy to imagine such a scenario).
But computers are nowhere near being able to have conscious thought. So what's the difference? Clearly it's not just natural language that differentiates the conscious from the unconscious.
They're only chaotic if you treat them in aggregate, which a superintelligence wouldn't do. It would be less lossy to get all the positions of the particles and figure out exactly what each one would do.
Something has to compute the universe, since it is currently running...