|
|
For advice on writing a Research Statement or Proposal for a PhD application, here is my brief advice. Please consider that this is just my opinion! The statement/proposal document is meant to be ~2 pages, and to inform the admission panel how you write (how closely you sound like an academic already). Opinions vary on what this should look like - google for "Phd research statement" to see the spectrum. I think ideally, UCL's CS Department and most other universities want to see your passion for an area (so the Intro should say why a specific problem is important and interesting) and how much you already know about being a researcher on a topic. So it should read a bit like a mini paper. Instead of results, you describe how you would validate the idea you're proposing, and what good vs bad results look like. It's crazy of course, because it assumes you've been in a field long enough to know the related literature and have identified a new problem and reasonable solution. So the strategy I recommend to people is to take 4-5 papers you know, pick or invent a problem related to them, and suggest a conservative solution whose outcome can be measured. This is NOT binding, need not overlap with the funded studentship, and it's very rare that someone does a PhD on the idea in their application. Though it should relate to Computer Vision/Learning as a field somehow. It's an excercise and a barrier-to-entry to separate class-takers from professional researchers. In your shoes, I'd devote some fixed number of hours to just get it done; I'd guess that the rest of your application has further strengths that also hopefully make the admission panel curious enough to want to interview you. And it's a monograph, so I don't pre-check these for applying students :) =============================================================== Postdocs No further funded postdoc positions are currently available in my group, but a joint project I'm part of in machine learning has one vacancy at present. ================================================================ PhD Studentships Sorry, my group is very big right now, full of brilliant researchers, and no other PhD Studentships are available at this time. ================================================================ I do not have space or resources for hosting interns, so please do not ask. If you want to apply to do a PhD with me at UCL: If I am currently hiring for a particular position, it will appear on my homepage and the section above. General unsolicited applicants, please read on. ELIGIBILITY: There is a lot to say, so please see the separate section at the bottom BEFORE reading on. GENERAL: (other options for PhD studentships at UCL) The Computer Science Department of University College London (UCL) is always accepting applications for EngD and PhD positions. A few specific slots are open now, and further ones will continue to open each academic year. Strong communication skills in English are essential. UCL's computer vision and graphics have expanded, with dynamic new research in several areas. We are looking for really driven individuals who have demonstrated talent or expertise in some specific way. Experience in visual-computing-related fields is ideal, but for example, there could also be a good match for mathematicians, physicists, roboticists, or those on the implementation side of HCI. Most of my projects have significant machine learning components, and we work closely with researchers in related groups and companies here at UCL and abroad. We are especially keen for applicants who really have one such area "nailed" and are ready to apply/expand what they know. HOW TO APPLY: To be considered, you must fill in the general UCL application form. Please see http://www.ucl.ac.uk/prospective-students/graduate-study/application-admission/ where you can download the forms and guidelines. Informal enquiries about the current projects can be made to me by email. Please put engdapp42 or phdapp42 in the subject line. Other emails will be ignored! (Yes, it's like a spam-filter) When emailing, please also include a CV, research statement, and contact information for references with a pointer to your Master thesis and publications in case you have any. Students without Masters are encouraged to apply to do an MSc in Machine Learning, Masters of Research (see the stand-alone program) or the UCL MSc in Computer Graphics, Vision, and Imaging. SECReT funding is another option, if you are interested in vision for surveillance or action recognition (and are a UK "resident"). SEE ALSO related Vision/Graphics openings (and check our general CS page for updates; http://www.cs.ucl.ac.uk/vacancies.html) ELIGIBILITY: "British residency" is required for many Doctoral programs in the UK; meaning UK citizens or someone from the EU/Swiss who is ALSO resident in the UK for the last 3 years (see 'EU Nationals in the UK' here). If you also have excellent academic qualifications, then you are in the best situation, in that pools of such funding seem to form every year. Put the word britapp in your subject. The key issue for UK and EU students is "fees-only" vs "fees and stipend". To get funding for students from other countries (including the EU) requires a proven track record. Essentially, you must have done some significant research of publishable quality already, before or during your MS degree. I list just a few links below, and apologize in advance if I don't respond to your request for a letter of support: I get so much unsolicited mail of this type that I frequently can not give an application the time and attention it deserves. It will help if you put the word phdapp in the subject. Please look here for the normal admissions procedure for UCL. Note that we have a separate (and excellent) MS program that is worth completing before launching into a PhD. If you are interested in research, try to be persistent - this is a hugely rewarding career! Get more info about funding if you're - American: Marshall
Scholarships Whenever I obtain funding that is unrestricted, I advertise the position with specific instructions here. |
|
Back to my homepage |