A while ago I asked a question about the influence of theoretical physics on those working on SAT solvers. Today I learnt that there’s a whole host of domain theorists working in the general area of general relativity (example). Apparently an area of mathematics developed to address the semantics of the lambda calculus is now being applied to the study of space time. What other branches of theoretical computer science, or areas of mathematics developed by them, are being used by theoretical physicists?
Would Wolfram’s genius - or nutty crankfest - about automata be an example? I think maybe there are particle interactions that work similarly. He did write a paper about antiprotons or something as a teenager, I think…
“At the age of 15, he published an article on particle physics[4] and entered Oxford University (St John’s College) at age 17. He wrote a widely cited paper on heavy quark production at age 18.[2]”
So says Wikipedia about Stephen Wolfram.
Check out the last chapter of Li & Vitanyi.
Ah yes, I forgot about Kolmogorov complexity.
I’m not sure Wolfram counts. His “new kind of science” is widely seen as a joke in computer science circles and just a rehashing of what was already known put in more bombastic terms. Besides, it isn’t clear to me what influence his ideas on automata theory had on his paper on particle physics.
Information Theory might count. Claude Shannon started it while working on compression and signal processing and it’s branched out into multiple areas.
Domain theory is not really CS although it presents itself as such. Panangaden (whom I know quite well) is a physicist by training working in a CS department. He is also extremely knowledgeable about mathematics, category theory in particular.
Somebody invented a vast new computer, by far the most powerful ever, and switched it on and asked it whether it would be able to tell us the meaning of life. It hummed for hours and made the lights dim and the room hot, and eventually replied that yes, it would be able to, but it would take a long long time, and it would have to invent Mathematica first.
I think it’s surprisingly difficult to come up with examples. Even areas like quantum computation and cryptography have been more in the way of newly created fusions than a flow of ideas in one direction.
Worth realising that theoretical physicists traditionally had zero formal exposure to theoretical computer science. In my case in the UK in the late Eighties, I had a couple of courses with experimentalists teaching us how to knock out Fortran code and that was about it, though one picked up a certain amount by other routes.
But Shannon was blatantly explicit from the start that he was partly using existing ideas from statistical physics.
Sure, but one would be hard pressed to come up with any theory that wasn’t based in part on something else.
Except that a) as I already noted, Shannon himself acknowledged the direction of the dependence and b) some theoretical physicists have gone as far as suggesting that Shannon was successfully attempting to exploit and co-opt the prestige of the existing ideas in physics. (Offhand, I believe it was Murray Gell-Mann who subsequently proposed the rule that, following the manner of Shannon, one should always dub something “entropy”, just because of the associated physics qudos.)
There have been times when the fashionability and origins of information theory has meant it has been attractive to physicists, but still Shannon presented it as being created in imitation of statistical physics.