1,222 reputation
1519
bio website
location Baltimore, MD (USA)
age 24
visits member for 2 years, 9 months
seen 6 hours ago

2nd year graduate student at Johns Hopkins University in Physics. I'm primarily interested in particle theory and mathematical physics. I have some interest in pure mathematics, especially geometry and topology, and outside of particle physics in other fields including cosmology, topologically protected phases in condensed matter physics, and information theory.

I haven't had much time for this site recently. It's rather difficult to use this site to ask research questions, and it seems there are many people here who are capable of and willing to answer the basic questions which are very common here. If someone starts a new research-level Q&A site about physics (either on SE or elsewhere) I'd be interested in participating.


Sep
21
comment What do theoretical physicists need from computer scientists?
I should add a caveat that I agree with Peter's original comment entirely, as I read it. I thought he was commenting the idea that "more powerful computers will always be available" is irresponsible. If I seem to support such a position, in fact I do not. That some of my colleges thought this way is part of why I moved out of computation into more mathematical areas. It makes algorithm analysis seem pointless, which it isn't. However, I also think that procedure and algorithm-based programming has its limits. Machine learning and data-driven theory are the next step when stuff hits a wall.
Sep
21
comment What do theoretical physicists need from computer scientists?
What I'd really like to see is if someone could come up with a list of commonly used algorithms in physics that aren't well known outside physics. That would be a post in the algorithm analysis area that I'd support. But I think such general things as "Improve your algorithms" or "Give us $O(n^2)$ matrix multiplication" aren't terribly good examples for this question. The former is somewhat irresponsible without having read the algorithm itself. How do you know that improving the algorithm will solve the problem better than machine learning? The latter is discussed in the previous comment.
Sep
21
comment What do theoretical physicists need from computer scientists?
Alright, I now understand what you're talking about, and I can see that you may be right. However, if that's what he meant with that comment, I think it's off topic on this site and this question. The question is about what theoretical physicists want particularly, not what everyone who does computation wants in general. I'd be ecstatic for $O(n^2)$ matrix multiplication, but I don't think that's a reasonable thing to ask. It'd be like asking mathematicians to answer the Hodge conjecture. It's already well-publicized and people are working on it, but it's not an easy problem.
Sep
21
comment What do theoretical physicists need from computer scientists?
I'm afraid I don't understand what point you're trying to make, then. I agree that algorithms can be improved, but barring a rather drastic and unlikely advance it wouldn't seem that programmatic computation alone will solve the issues posed. Are you disagreeing with me here? Or are you disagreeing when I say that using machine learning techniques may help alleviate the issue? I'm finding myself agreeing with what you say, so I'm not sure what I said that you're disagreeing with. I suspect it may be a misunderstanding. Could you give a quote to which statement I made that you don't agree with?
Sep
21
comment What do theoretical physicists need from computer scientists?
I'm well aware of the problem of optimizing matrix multiplication, but I'm not convinced this is something that is specifically or especially wanted by physicists. Virtually everyone in the world who does any computations wants to multiply matricies faster. Of course if someone can find a faster algorithm for matrix multiplication, singular value decomposition, fourier transform, etc. then it would greatly benefit physicists, but that's fairly obvious. Also, at the time I posted this answer, Aaron was asking for specific examples; hence, I tried to stay fairly concrete with the problem.
Sep
21
comment What do theoretical physicists need from computer scientists?
@Peter After reading my writing again, I can see what you disagreed with. I removed the second paragraph, which was saying that algorithm analysis is not useful. I meant this only in the very narrow context I was dealing with, and even then it wasn't well explained. I still think that faster algorithms wouldn't solve the problems posed here, but of course they would help.
Sep
21
comment What do theoretical physicists need from computer scientists?
@Joe: Yes, but in the very next sentence he asks about practical problems, which was to me an indication that he was looking for answers from both the theoretical and applied points of view. Also, to clarify, I was not taking computation and computer science as synonymous, but I do have to admit I don't know much about what theoretical computer science consists of beyond, say, Hopcroft & Ullman. The theoretical computer scientists here do research in things like biological computing, which is, as far as I can tell, not what most people mean when they talk about theoretical computer science.
Sep
21
comment What do theoretical physicists need from computer scientists?
While I certainly like this answer, I don't find any indication in the question that Aaron is intending for it to be mostly directed towards the area of theoretical computer science; he even asks about "creation of practical tools". Of course the questioner is a theoretical computer scientist, but from what I can tell the question is equally about all branches of computer science.
Sep
21
comment What do theoretical physicists need from computer scientists?
Also, let me add that it's great to see such a notable researcher on this site. If you could write up your own experiences in greater detail as an answer, I'm sure it would be highly appreciated by everyone here.
Sep
21
comment What do theoretical physicists need from computer scientists?
While I can't disagree that it would be nice to speed these up, considering the amount of work that was put into optimizing every aspect of the particular codes I worked with (by professional programmers as well as physicists), I'd be surprised if they could be sped up more than linearly. While a speed up of a million times would certainly increase the quality of the simulation, in this case the system is intrinsically chaotic and even a million times more computation wouldn't get you that far. In such complicated systems, data-driven theory seems to be the best way to approach it.
Sep
20
comment What do theoretical physicists need from computer scientists?
I mentioned machine learning in a broader context, see for instance the last sentence. I can't claim to be knowledgeable of anything in the field except a little bit on neural networks, which is why the answer almost exclusively discusses them. I would agree that there is much beyond neural networks to be gained, but I don't know quite what it is.
Sep
20
comment What do theoretical physicists need from computer scientists?
Alright, I must admit that this doesn't exactly fit your comment on Michael's post, but I think it's a start.
Sep
20
answered What do theoretical physicists need from computer scientists?
Sep
16
awarded  Scholar
Sep
16
accepted Why should one expect closed timelike curves to be impossible in quantum gravity?
Sep
16
accepted How Fundamental is Spin-Orbit Coupling to Topological Insulators?
Sep
16
comment How Fundamental is Spin-Orbit Coupling to Topological Insulators?
I see. I probably misunderstood because I'm not coming from a condensed matter perspective and Kramers theorem is not something I was well-acquainted with. I think this direction should be sufficient to begin reading on the subject, and hopefully correcting my misinterpretation.
Sep
16
accepted What are the justifying foundations of statistical mechanics without appealing to the ergodic hypothesis?
Sep
16
comment What are the justifying foundations of statistical mechanics without appealing to the ergodic hypothesis?
Alright, after rereading it seems to me that you are claiming that the standard foundations of statistical mechanics do not invoke the ergodic hypothesis at all, and that the emphasis on it is the fault of poor pedagogy rather than a bad choice of fundamental principles and postulates. I misread that earlier. In any case, this fully and completely answers the question, so I've accepted it.
Sep
16
asked How Fundamental is Spin-Orbit Coupling to Topological Insulators?