j3doucet
-
Posts
34 -
Joined
-
Last visited
Everything posted by j3doucet
-
School ranking really depends on your interests. Waterloo is great and all, but in game AI research, for example, Alberta is one of the best in the world and Waterloo barely studies it. At the graduate level, who you work with aught to be a bigger concern than where you will work. Unless you're doing a course based masters... Also, most Canadian schools don't use GRE. It might help, but it's certainly not required. Anyway, you really need to provide more details.
-
Maybe I'm way off base on this, but I tend to lend more credence to people who include such things in their email messages, rather than less. Maybe NSF is more random than the Canadian equivalent, but the impression I have for both awards is that the people who hold them are, by and large, solid researchers. Now, granted, there are researchers who do not hold national fellowships who are also strong, but an NSF or NSERC award is an easy signal to me that you've been vetted, and that I shouldn't worry to much about your competence if I want to work with you. Putting it in your signature is definitely going to make me more likely to view you favourably. Now granted, I both hold such an award and do _not_ mention it in my signature, but maybe that's a good thing given what I read here. Just wanted to point out that there are people out there who don't view it as pretentious at all, and who are glad to see it.
-
WOW! I got Admitted and screwed at the same time.
j3doucet replied to sdsualumn's topic in Decisions, Decisions
Well, lets look at the math. The degree costs you $50,000. On top of that, you won't be working. Don't know how much you make now, lets say you're making $Xk year. So you miss out on $X,000 for two years. So the final cost of the degree is $50,000 + 2*x Okay, so how much more will you earn with the degree? You need to earn both enough to offset the interest on costs, plus the interest you could have made on deferred income (let's say it's 5% a year), _and_ enough to pay down the principle. So, conservatively, if you want to pay off the debt from this degree in just say, 20-ish years, you need to make 0.07*(50,000 + 2*current_salary) more per year. I don't know a lot about salaries in sports management. Will they be 15-20% higher than what you're making now? If not, I'd say this is a bad idea. On top of that, student loan debt is the very worst kind you can get. You can never default on student loan debt, which means you could be paying it until you die if things don't work out. -
TakeruK has the right of it. A friend of mine likes to say that base pay is just enough so you don't starve (though of course, this depends on where you live).
-
I was seriously considering it, so probably others exist. If I had gotten off that waitlist at Cornell...
-
How to prepare for research in machine learning?
j3doucet replied to luffycullie's topic in Computer Science
Linear modelling is very useful as a foundational course. If you're interested in neural networks, much of that community has moved into support vector machines, which are very mainstream. Probably the best thing you could do though would be to get some experience working with real world data, and with the machine learning system's you're interested in. There are a number of repositories of free data (e.g. http://archive.ics.uci.edu/ml/). Go there, download a likely dataset, and then implement a simple back-prop neural network or the ID3 decision tree learner. You can compare your results with those generated by Weka (http://www.cs.waikato.ac.nz/ml/weka/). The theory you could learn in classes is all well and good, but unless you just want to do pure theory, the application details are going to matter more. -
Thanks! Suggestion 3 is especially clever. I hadn't considered that idea.
-
Something no one here seems to be mentioning, that I'd love to hear suggestions on, is dealing with travel reimbursement. So here's the racket if you haven't encountered it already: You need to go to a conference. The university says sure, no problem, we'll pay for it. Then they tell you to buy the tickets and hotel (2 or 3 months before the conference), and that they'll reimburse you after you fill out paperwork with boarding passes and receipts when you get back. Usually this takes another month to work its way through the system, so you're looking at potentially several thousand dollars worth of interest free loans to the university for 3-4 months. I usually end up putting it all on a credit card, getting advances for whatever the university will cover before the trip (e.g. conference registration), and then watching my bank account dwindle to zero and my CC debt climb toward its maximum for a nail biting couple of months. For example, this month I have $200 cash on hand, $1,500 in remaining CC debt, and $1700 in loans to 3 different universities (admissions visits and a conference). Often these are hard to budget for too: it's not worth living on pennies because you _might_ get into that conference in Japan later this semester and thus _might_ have to shoulder a $1500 loan for a couple of months in there. Any suggestions? I've thought about trying to get a line of credit for it (lower interest rates), anyone done that?
-
Sorry, I should clarify: The CS department is offering me an extra (CS only) award which covers my tuition, and a full TA. My supervisor is suggesting an extra RA stipend on top (though reduced). So it depends not only on your school, but on your department, faculty, and supervisor as well.
-
It depends on the school. Waterloo is quite generous, and will cover my tuition plus some more. I imagine other schools will at least wave tuition, but some schools might not be willing to do this.
-
Given your background, you might take a look at the subfield of bioinformatics. I have friends who came there directly from a biomedical degree without any programming experience. Naturally you'd have to learn some, and most schools would require you to take some remedial CS courses as well, but it's not impossible to jump into a graduate level CS degree from biomedicine.
-
I think you should do a Canadian masters. You'd be a good candidate, and research-based masters students are pretty much all funded. Also, even with the international fees, our tuition rates are quite reasonable. Check out University of Toronto, University of British Columbia, and University of Waterloo.
-
My understanding is that this means no public, broadcast announcements. I doubt they are expecting anyone to keep this from their immediate family, or even necessarily from their supervisors (hlpes them plan next year's budget), but they clearly want to keep the press release to themselves. If you make a public broadcast anywhere the press could get hold of it, my understanding is that you're violating the spirit of the statement if not the letter as well. That said, I am not NSERC. You could try emailing them. Also, I suspect I'll run into you in person at some point, fellow Warrior. Congrats.
-
Best CS reputation in Canada, UBC or UWaterloo
j3doucet replied to Deema's topic in Decisions, Decisions
Disclaimer: I'm at Waterloo now. The school is pretty great. Some of the main selling points: - Great funding: Many 10k/year merit based in-course/entrance scholarships for CS grads. No 'salary cap' (Some students make $45 or $50k/year by holding several different awards at once). National scholarships are 'topped up' with an extra $10k/year that stacks with the CS specific ones. Mix this with the very low cost of living, and you can be paid a handsome amount. - The school is ambitious. This doesn't seem like a big deal until you experience it. The place I did my undergrad was 200 years old, and seemed content to be just another mid-range research university. At 50 year old Waterloo, people (faculty members) talk about trying to become an Ivy League level school. That's their goal. They go far out of their way to improve the school's reputation, and it's really on the rise in that regard. The marketing department wins international non-academic awards. This 'buzz' can be a great motivator. -Students own all their IP. If you invent something, you own it. The university also has patent support though, where they'll help you patent something in exchange for a cut. That said, the town leaves much to be desired. It's close to Toronto though, so many students go there for the weekends. -
I'm on a wait list for PhD admissions at Cornell. Apparently results were supposed to be out April 6th, but I haven't heard anything as of yet. Is anyone else in a similar situation? Has anyone heard anything?
-
There are also quotas by school, and some of them were close to being exhausted (they get a quota for a 3 year period, and this was the 3rd year). That might be another reason...
-
Perhaps they count from 0 (i.e. maybe the committee is headed by a computer scientist...)?
-
I got it, rank = 38. So 38 < cutoff < 55
-
All true, but some schools, and some departments, do not have dedicated clusters (or have ancient ones).
-
If you're going to be doing any kind of heavy lifting at all (i.e. simulations), then you need a desktop. Laptops do not like the sustained heat output that comes from using 100% of the CPU power for days at a time. I have personally seen 3 laptops die from this sort of use.
-
It won't prevent you from publishing or researching it later, but you might get scooped. That is, if you really do have the kernel of a promising research idea, then someone else might see it, and do research on it now. One way around this is to publish the basics as a technical report. Typically, you can register any technical document with your University library or a similar service. It gets an official registration number, and you retain copyright. Then post the tech report to an open-access archive like arXiv.org (this is free), which will result in it being disseminated throughout the scientific community. Once you've done these things, make your GitHub project, and put it under a citation based licence, with a pointer to the tech report. Now, if anyone wants to expand upon your work, they have to cite your tech report. If no one does this, you have a nice thesis topic waiting for you. If it turns out lots of people are interested in it, your tech report can get a large number of citations, which (if you market it appropriately), can be worth as much as a top conference paper.
-
Research is pretty important. You should volunteer with a lab. Any lab, doesn't have to be absolutely the thing you want to do. Practical research experience opens a lot of doors... Many professors are happy to take you on as unpaid help in your undergrad. Initially, you might have to do grunt work - data entry, routine coding. While you're doing this, spend as much time as you can talking to the other lab members. Get a desk in the lab if you can. Once you know the basics about the area, start reading papers, and soon you'll know enough to get a real research project.
-
To clarify, the department I want to study in is very highly ranked, though the school has a good overall ranking as well. In fact, the department is even in the top 10 in the subfield I want to work in.
-
GRAVE CONCERN about my research publication
j3doucet replied to jamundergrad's topic in Computer Science
If the paper has not been published (that is, you withdrew it from the conference after it was accepted), then you can definitely publish it elsewhere. Unless you signed over copyright, you should be fine. If you are worried about it, you could ask the program committee chair at the conference you want to submit to next whether this would be a problem, but I can't imagine it being an issue. EDIT: I misread your earlier post. You'll have a very hard time (if not impossible time) publishing the paper elsewhere if it's already appeared in the WORLDCOMP proceedings. However, many venues will accept papers which contain as little as 30% original content (if the other 70% are from a paper by the same authors), so there's a good chance you could expand it a bit and publish it at a better conference. -
How bike-able is Waltham? Are there lots of cyclists? Are motorists courteous? Also, as an international PhD applicant, I was quite pleased with their financial aid package, although the cost of living seems a bit high.
