Jump to content

ZNtheory

Members
  • Posts

    17
  • Joined

  • Last visited

Everything posted by ZNtheory

  1. In case it is helpful to give advice, here is what I have learnt during my undergrad Mathe Courses: Prob Theory (based on calculus) (A-) Numerical Analysis (A-) Linear Algebra (two terms) (B+/A-) Mathematical Analysis (three terms) (A/A/A-) Real Analysis & Lebesgue Measure Theory (A) Functional Analysis (B+) Complex Analysis (A) Topology (A) Stat Courses: Multivariate Stat (A) Time Series (A-) Regression (A) Math Stat (A)
  2. Hi everyone, I am a first year master student of applied math at Penn and want to apply for Stat PhD next year. I am posting this to ask what course I can choose so that I can maximize my advantages in the application. The course I have decided to chosen is Numerical Linear Algebra (Applied Math PhD core Course) Probability Theory (Wharton PhD Core course) I can still choose 1-2 courses and I have a candidate list Mathematical Stat (Wharton PhD core course) Probabilistic Machine Learning (CS PhD course) Topic in Harmonic Analysis Analysis (PhD core course) I know the second one is a bit different, but I really love the content of this course and I actually had some research experience related to what this course will cover. For the first one, since about 70% of the contents are covered in my undergrad math stat course and I do a very great job on that course (97/100), I am a bit hesitated about this one. The other two covers deep things in analysis which I am really want to learn. Any suggestions are appreciated! Thanks in advance.
  3. Another question is that what is the best thing that can boost my application for PhD in 2023 fall. I am now thinking taking some advanced math course such as functional analysis ( though I have done this during my undergrad), probability theory etc which are courses for math PhD. But another way might be doing a theory based project and result in a nice research paper. I wonder how to balance these in a not very long master program...
  4. Hi guys, I am trying to update this thread and having some more questions. I have accepted the choice to UPenn. Moreover, my project during the undergrad is on arXiv and submitted to EJOS (theory & methodology paper). I am a bit curious about the reputation of this new journal among the statistician? Will this be a good choice for a ML theory paper?
  5. Thanks for this! I think they do offer TA to give financial aid. I also contacted the chair of the program and they said they accepted several students from the master program into their own PhD this year as well as last year. I think this is a very good result since only half of students (5-6) will go for a PhD after finishing the degree.
  6. Hi Daniel, I think these students might stay at other departments apart from math but some very "close" departments. At least the program website told me this way. Or maybe this is just data for one year.
  7. Thanks for your reply. As for AMCS program, it is said that they accepted around 10 students for this master program. This program is designed for future PhD study. One can choose any PhD level course including Wharton Stat courses. As for research, I heard from previous students that they could usually find someone to work with and around half of the graduate stayed at UPenn. For those who applied other PhD programs, it is said that people went to place like Columbia EE or UPenn EE.
  8. Hi everyone! I am posting this to ask for some suggestions on the school choice. I just got UPenn AMCS master and JHU AM&S PhD offers. Out of USA, I have got the pure math master offer from Warwick and am still waiting for news from Oxford, EPFL, which are both pure math master programs. My ultimate goal is go into academia and my own research interest is in theory of ML and statistics. Could any of you give me some suggestions on how to choose these programs? For more information: my undergrad is not math but economics and statistics, so I don't think I have that solid background compared with math major student. However, apart from usual statistics courses (mathematical statistics/Probability), I have taken most of the analysis courses including Mathematical analysis, Real/Complex Analysis, Functional Analysis besides usual Calculus. For algebra, I have taken linear algebra for two terms. Also I have taken many computational courses like Numerical analysis and Optimization. I really want to do theoretical work in future so I know I need to do more... Thanks in advance!
  9. I once was supervised by some researchers at UCL on the topic of Gaussian Processes and its application on uncertainty quantification. I think many researchers at least at UK institute has done research in relation to UQ as well as its application in machine learning though many of them hold their position in Stat Department. As for HD statistics, I think it has been popular since the sparse models have been widely used and developed. More recently, HD statistics have been used to more broad aspects like analyzing the generalization of overparameterized models. Therefore, I think whether research on either HD statistics or UQ should be fine if you want to secure a job at academia. Just personal thought. :)
  10. I think Prof. Weijie Su who is an AP in Wharton in UPenn has started to do research in the area of fundamental of deep learning and published some papers on conference like Neurips. I think you should check him in case you are interested in his work.
  11. I looked through the website of Department of Statistics in UBC and found that some very strong guys have joined who may focus more on machine learning. I think you are probably right to get into a department with people whose interests are diverse if you are not so stick to Casual inference.
  12. Thanks so much! What level UCL you think should be in? Match or Safe?
  13. Thanks for comment! Yeah, I am the first author for all three papers. The reason I think my letters are very strong is that the supervisor in UCL as well as the paper collaborator has told me that he would really love to take me as his PhD student if I want to apply for UCL. The other two LORs I think will be strong is because these two professors have already written me a letter when I apply for REU at MPI this summer. The faculty in MPI read the letters and give me very positive feedback. Which safety schools you recommend which have professors working on theoretical ML especially in optimal transport can fit me well? Your comments are always appreciated, bayessays.!
  14. Thanks bayessays! I just make a list. Could you guys help to comment on them? Thanks so much. Reach: Stanford(stat) MIT(machine learning) CMU(machine learning and stat) UCB(stat) Princeton(ORFE) U of Chicago(stat) Upenn(Wharton) NWU(stat) Match: PSU(stat) Michigan(stat) UNC(stat) Minnesota(stat) Safe: TAMU OSU UIUC I want to add some safe schools, what do you suggest? Thanks again.
  15. Thanks for the encouragement! Although it is very very unlikely for me to be admitted to Stanford, I will try my luck.
  16. BTW: my main research interest is about computational statistics and theoretical machine learning.
  17. Hi everyone! I want to apply statistics PhD program for 2021 fall. I am posting this to want to know where I can roughly aim for. Undergrad institution: One of the best schools in China Exchange institution: one of G5 in UK Major: Statistics GPA: 3.7 Major GPA: 3.83 Background: International Asian male Relevant math courses: Calculus 1-3, Mathematical Analysis , Linear algebra 1-2, Probability, Mathematical statistics , Numerical analysis, Real analysis(measure theory), Complex Analysis, Functional analysis, Stochastic Process, Time series I got all A or A- except for linear algebra. Will take: Linear Regression, Multivariate Regression GRE general: Quant: 166 Verbal: 153 Writing: 4.0 TOEFL: 103 Research experience: Mainly 2 long periods of experience and result in 3 papers. The first experience is when I am in my sophomore year, I spent whole summer in McGill University and worked with a Professor there and another Professor in U of Connecticut. It is about computational statistics which mainly focus on developing a GLM model with extreme value response(heavy tail) and develop a very fast proximal gradient algorithm and write a R package. The second experience is during my stay in London, I work with Professor in UCL and mainly focus on theoretical machine learning, which results in two papers. The first is about doing fast inference for generative model using QMC method. The second one is about deriving a Sobolev bound and L_{\infty} bound for the derivative Gaussian Process approximation. Papers: 1. The first paper finished when I was in McGill will be published on JCGS. 2. The second paper which is about QMC will be published on JMLR. 3. The last will be submitted to Neurips 2020. LOR: Three Professors I have mentioned above will write me a LOR respectively. They will be very strong since all of them have given me quite positive comments and I have worked with them for a long time and collaborated with them on some papers. I am taking up math GRE this october. Any suggestions and comments are appreciated. Thanks!
×
×
  • Create New...

Important Information

This website uses cookies to ensure you get the best experience on our website. See our Privacy Policy and Terms of Use