zer0

I have accepted the offer from Georgia Tech. It is the ACO program from CS Dept. Looking forward to it!

I am not familar about Caltech coz they're so small. I didn't apply UT Austin but I did some survey before. The theory group there is quite strong, but mostly are complexity/crypto people. So if you like either one of the two, that's definitely the place to go. I got accepted by GaTech and decided to go there coz they have a very large group that covering almost every subfields of theory.

Mainly depends on your interest. Which subfield you like in theory? If combinatorics style (like network design, graph algorithms, etc), Maryland is definitely the best in these three. And they just hired a very strong one Mohammad Taghi HaijAghayi recently.

As a master international student who also applying theory, here is my 2 cents Math/CS double major definitely boosts your chances of getting in, but, as far as I know, several students in US college who do theory already have papers (some even in the top conf). And you should know that the slots for theory students are not so much. It's not like other fields where a prof can have about 10 students. In theory, a prof mostly has 2 or 3 students, sometimes even 0/1. Congratulation for the offer from Toronto. They have a strong theory group. Also do you have any particular subfield you want to do in theory? From your list, it seems to be complexity?

MPI is Max Planck Institute, which is one of the best places for theory in Europe.

+1 for AGT, my school list: CMU, Cornell, UW-Seattle, Gatech, NYU, Northwestern, UPenn, Maryland Dartmouth, Rutgers, USC, MPI (my advisor has a connection there) We already have 3 people working on AGT in this thread. Events with low probability DO happen (or there are really so many people working on AGT right now?)