the poisoned pawn

Some of the ranking. You can google "computer science conference ranking" for more. Usually, people do agree on the tier of the conferences. http://www.ntu.edu.s...ourav/crank.htm http://webdocs.cs.ualberta.ca/~zaiane/htmldocs/ConfRanking.html

Thanks for your responses. I have an impression that I will have to give it more time. But it's only my first year, change can still be made, but with caution.

So here's the case: I'm in a CS lab with around 6-8 people in it, and I'm sharing an advisor with 5 other students. So my advisor is pretty busy splitting his time to advise those other students and to do his own stuff. In my research field, I'm leaning more toward the mathematical and theoretical side which requires a bit more interaction with my advisor (to get more ideas, resolve math confusions). But in our individual meeting, which is already short, he is short of talking about high level things and he does not really go deep into the questions that I have. I was expecting a closer work with my him and being in a smaller group. Is it a normal thing to concern about? For some reasons, I guess that having an advisor (who has fewer students) that can work more closely with you on the research will be more efficient. Is it true for your case? I'm a first year student. So this may just be my feeling.

I will think about sharing mine but here's a suggestion for a good SOP that stands out from the rest. Instead of going from your background to your research interest, do the other way around. Be straight to the point at the very beginning. 1. Your very brief introduction (education: where are you from, which school did you attend, your major and your goal to pursue a graduate program in CS and AI) 2. Go straight to your interest in AI (with some brief explanation: why you like it), some projects and aspects of it that you want to study and do research on, which professors you want to work with. (Also add a compliment about that school's CS department and AI program) 3. Explain what makes you think you and that school will both benefit from each other: now, explain in depth about your academic standing/achievement, RESEARCH EXPERIENCE. 4. Make a short conclusion.

You definitely have a good shot at the top 20. AW is not that important, I got 3.5 and still get in a top 20 as well. The most important thing is you are able to find professors that have a good fit with your research interest and mention that in your statement of purpose. Many students are not admitted not because they are not good enough for a certain school but because none of the professors finds them suitable for his/her current research. Of course, you should apply to some schools outside the top 20 as well.

If you think research is as simple as homework, midterms, etc that you can fully prepare before even coming in, then you're quite wrong. It may turn out that research takes a lot of paper reading, guide from your advisor to progress even if you come in prepared. For the working part, of course, they key point is not to get caught, but you can say the same for bank robbery, cheating during exam, stealing light bulbs in Walmart, etc. In any case, hope your company pays for your part time MS. If not, you first need an admission to execute this plan. So, good luck.

Should be a good LOR if the prof you work with write you a good letter. There are a lot of PhD CS students with background of cognitive science, linguistic, math, engineering, etc.

If you're a U.S citizen, you can go out and work for a company 1-2 years, most of them will pay for your masters tuition. If you're an international student, you can seek for scholarships, or work for several years to save up money for the MS program. For the trick that you're trying to do: you're not the only one who does that. I know a lot of people who do so which make me really pissed because it kills or severely reduces my chance of getting in schools that I like. But I will answer part of your question here: for MS/PhD program, you have to work as a TA or a RA in order to be paid. If you don't contact with any professor and don't do research, then you have to work as a TA (will consume a lot of time). Unless you have a fellowship, to be funded, you have to work 20 hours for the department.

I double major in math and CS and also took the math gre (though I didn't report the score), so I can give you some advice here. Math GRE will surely help your application a lot especially your field of research is close to theory. But prepare for a lot of topics like topology, real analysis, abstract algebra, ... Math GRE does not require a deep knowledge in any particular area but it will cover a lot of areas (in other words, a little bit of everything). A lot of these topics are mainly studied by math majors. So, studying for it will take a lot of time. I would suggest you put research in a higher priority than the Math GRE. If you take a lot of math courses and have good grades, it is already a proof for your math's maturity.

1: NP hard 2: NP complete The rest: no clue

Based on what you posted, yes, I think you do have a chance for a Masters admission at those schools. My friend has a pretty insignificant profile and got admission into the Masters program of Cornell, but he got rejected from all the PhD programs he applied to.

I'm also an international student, not from India, Japan or China though. As far as I know, if the school you apply to have some professors/grad students that are from your country then probably one of them will be consulted when evaluating your transcript. Saying that, there's pretty much no need to convert to 4.0 though. It may not make sense and not reflect the equivalent performance.

I believe it is possible to have a research position in industry without a PhD. However, if you are really passionate about research a PhD is a better choice because of the extra 2-4 years of sole research. I know some people working as a researcher without a PhD, but then decide to come back to get a PhD as they feel the need to do so.

It probably won't matter for schools in the top 20-25, but the story could be different for lower rank schools. In any case, GRE is the least important among GPA, research experience, SoP and recommendation letter. Don't get a too bad score because it may raise some red flags, but I think your score is fine given that you have a decent research experience, recommendation and a good academic standing.

No. There's no need to, or I can say it's not a must. You can try to email your favorite professors and ask about some of his current work and share with him/her your favorite area. That may do the trick. They don't have time to reply to a lot of emails. If you have time to have a reasonable look into the projects/papers of the professors that you want to work with and mention how it matches with your background/interest in your SoP, I think it is just as good.

I'm not sure about that though I'm an international student (my undergrad in the U.S). Applying to 30 schools is just insane. I don't think it's necessary or worth it at all to apply more than 15 schools. So far, when I check out the incoming PhD student list at my current school and my future school (both in the top 30 and 20), there are 50 or 60% of American students and 50-40% internationals. Even though, of course, getting admission is tougher for internationals, it is much wiser to optimize your applications rather than mass apply. My friends back home apply around 10-15 schools and if his/her profile is rather ok, I think that person may be accepted to 2-3 schools.

I think a specific ranking like 7th, 12th, ... is rather meaningless. The tier is the better term. Let say MIT, CMU, Berkeley,.. are in tier 1, UCSD, UMass, UMich, .. are in tier 2, OSU, UCI, Chicago, ... are tier 3 and so on. So, pick several schools that fit your research interest the most in each tier and apply. That was how I did.

dress nothing

MS/PhD at UMass Amherst.