« first day (3766 days earlier)      last day (1148 days later) » 

6:13 AM
Hi,

Can provide me any resources for learning machine learning / neural networks in Python for applications in Cosmology ? I have zero experience with the topic, so resources which start from vary basic and then slowly build up to solving basic problems in research would be very useful. Particularly I need to learn Artificial Neural Networks (ANN) to apply in Cosmology problems.

Thanks in advance.
 
 
2 hours later…
8:17 AM
Is “Group Theory for Physicists in a nutshell” by Zee a good book for self-studying group theory?
 
 
2 hours later…
10:00 AM
@john Depends - if you really want to understand representation theory (which is really what physicists mean by "group theory"), I believe you have to learn it from the mathematicians, but if you want to get the kind of "working knowledge" most physicists have, then books like Zee's are fine.
 
 
1 hour later…
11:02 AM
Hi all, how do you guys read physics textbooks? I find the process of stopping at every equation I don't understand and deriving it before moving on very tiring and slow. Any tips to share?
 
fqq
11:17 AM
it really depends on the book and what you want to get from it
 
12:06 PM
@ACuriousMind thanks! Do you know a better book? I’m interested in the math, but not extremely rigorous, and Zee’s seems to me the best option
 
@john It's been a while since I studied the subject, but Georgi's book is apparently a classic, though I personally wasn't a big fan. I rather liked Lipkin's "Lie Groups for Pedestrians", was quite a nice book even though the notation was a little weird. But all in all I found Zee's book to be the best, for precisely the reason ACuriousMind mentions: it gives you a working knowledge rather than an in-depth understanding.
 
12:34 PM
@Philip One of my professors also mentioned Georgi’s. I’ll look at both of them
 
 
1 hour later…
1:53 PM
@john I hold that Georgi is terrible, and sad that such a pioneer could not write a more formal presentation of the basics of his work.
 
 
1 hour later…
3:05 PM
grrr...some Outlook update has changed all the custom colors in my calendar, first time I'm actually angry about the proverbial moved cheese :P
 
3:23 PM
@ACuriousMind these computer programmers: they think they know better... ;)
 
3:37 PM
can someone help me with this limit?
How do I justify that this integral in the limit t \to infinity behaves like e^{-imt}?
 
3:52 PM
@ZeroTheHero I am a computer programmer, but you're not wrong :)
 
4:04 PM
@ACuriousMind I knew that...
If I had not there wouldn't been any fun in making the comment... ;)
 
4:23 PM
programming would be so easy if it weren't for all these annoying users
similar to moderating, really :P
 
D:
 
spoken like a true university administrator: things would run so much better with no faculty.
(and no students)
 
in that case I think the faculty often thinks the same about the administrators...
 
... and faculty of course think that things would be much better without administrators.
 
4:45 PM
Is BRST quantisation considered the strongest (in the sense of being the most rigorous) approach to quantisation in QFT?
 
@Charlie you can do BRST at varying levels of rigour like everything else in physics, it's just the method you have to use for (non-Abelian) gauge theories because "naive" canonical quantization can't deal with them.
 
Oh ok that's interesting
 
hey everyone
Is there a way to organize stars in a room?
 
Unsure what that means
 
vzn
5:02 PM
@DarkVader hi are you a grad student? what is your background? ML in physics is still new. wrt your query general data science refs are a not bad place to start. think cathy obrien is a standout author in the field. looking into cosmology datasets is another direction to go in, someone else was askinga about that recently...
 
I think if we're talking about gauge theories in general, the BV-formalism is the goddamn cannon
 
vzn
Feb 2 at 18:17, by M.N.Raia
Hi everyone! Hope you are all safe and healthy.

So. my advisor wants me to do the following research project: 1) Get a light curve from a astrophysical object like a quasar 2) Apply a algorithm to obtain the periods of this curve
 
5:49 PM
@Charlie in my room I discuss programming and physics with John Rennie sir
my C++ exam is coming up, and some of the stars are quite useful in my room
 
Ah, that kind of room/stars, what do you mean by "organise"?
 
I was wondering If I could segeregate the stars into "physics" and "C++"
 
I think you can bookmark sections of your conversations and label them, maybe that's better
 
6:27 PM
@john yes, maybe give these a try for an idea of what it's like:
 
 
3 hours later…
9:01 PM
Probably something I should write up into an actual question, but maybe I'll probe it a bit first here.. when you have a superconducting metal, far below Tc, are almost all of the electrons then paired up into cooper pairs? I find that hard to imagine, with superconducting gaps being ~meV, but Fermi levels in metals being eV
 
fqq
9:33 PM
@user129412 I am a bit rusty but I think I recall that only electrons "close" to the Fermi surface have an effective attractive interaction, and therefore can condense into cooper pairs. So there is no complete condensation even at $T \to 0$. But it's probably worth asking the question.
Unfortunately I never studied superconductivity/BCS very well. It's an embarrassing gap in my knowledge of cond-mat.
 

« first day (3766 days earlier)      last day (1148 days later) »