eric brochu | haiku factory

So I just found out that not only was our poster a finalist at the SIGGRAPH Student Research Competition, but we actually won FIRST PLACE! Which comes with a plaque and some cash! See what happens when you go to the awards ceremony at the wrong time and then get annoyed and leave because you assume you didn’t win anyway and there’s free booze somewhere else?

This is a really great (and really surprising!) award! SIGGRAPH is one of the most prestigious conferences in computer science, so winning the student research competition is quite an honour. And not one I ever expected to receive: to be honest, I was plenty pleased just to learn we made it to the finals. I never for a minute expected to win. And even though as first author and presenter, I get my name on the award, it belongs just as much to my co-authors Abhijeet Ghosh and Nando de Freitas.

(And yes, I’m aware how cliche that last paragraph sounds. There you have it.)

The prize money will get the three of us one fine, fine dinner at Tojo’s.

SIGGRAPH, originally uploaded by Mister Wind-Up Bird.

Well, I’m back from SIGGRAPH, probably the coolest Computer Science conference around. And also, I’m tired.

While Graphics tends to catch a lot of flack (not least from its practitioners) for being somewhat hacky, I did find people were generally more open to ideas and less ivory-towery than I’ve seen at some Machine Learning conferences. And there is definitely a subset of the graphics community that is very interested in Machine Learning, though there’s maybe a bit of a sense of casting-around on both sides trying to figure out just how they can/should be combined. I managed to persuade a fair number of people to listen to my poster spiel and give away a bunch of business cards, so hopefully something will come of this.

Highlights of SIGGRAPH!

• Seeing old friends, like Andy and Pinar and Tamy.
• Having our poster make it to the Student Research Competition finals. I don’t know which of the finalists actually won, but I’m pretty sure it wasn’t us. But really, it’s an honour just to be nominated. No, really.
• The Electronic Theater, which also involved watching graphics-godfather Jim Blinn play an original Asteroids game in front of a couple of thousand attendees.
• The quality of the presentations. It is extremely high by computer science standards. Everybody finished on time, and I don’t think I saw a single slide of just equations! I was even able to understand things that weren’t my exact field of research! Plus, there were sometimes questions from the audience that weren’t “why didn’t you cite me?” and “why didn’t you cite my student?”
• Seeing Peter Jackson’s new short film in the Animation Theater.
• The Emerging Technologies center, where, among other things, you could check out E Ink, Microsoft’s version of multi-touch interfaces, and, most awesomest of all, a 360-degree display.
• Scott McCloud‘s talk about comics and what makes them a unique medium.
• A mere five-to-one male-to-female ratio.
• San Diego. I knew almost nothing about San Diego before going there, but it’s actually a really nice city. Lots going on, perfect weather year-round, and a mixed commercial-residential downtown. Actually, it reminded me a lot of Vancouver, but with better Mexican food, and without the eight months a year of grey and rain.

As previously promised, I am headed to SIGGRAPH! I will be presenting a poster of my research with Abhi and Nando on Monday and Wednesday (booth I03): Preference Galleries for Material Design.

This is the first time I’ve publicly presented this particular research, so it will be interesting to hear what people think of it. For those of you who might have gotten here looking for information about my work, I can offer you a selection of the following fine links today:

• poster PDF [4MB]
• abstract PDF
• my other publications
• my collaborators: Abhijeet Ghosh and Nando de Freitas

update (aug 7): Our poster has advanced to the final round of the Student Research Competition! Pretty cool, but very surprising (to me, at least), considering it’s a bit of an odd fit for SIGGRAPH in the first place.

update (aug 22): I only just found out that we actually won first place! Awesome.

It took me two years to find an area of research, two months to turn it into a research topic, two weeks to write a proposal, two days to write a proposal talk, two hours to defend it, and two minutes to pass it. But now, I’m a PhD student no more: I’m officially a PhD candidate. With all the rights and privileges thereof. Which is to say: none, other than calling myself a PhD candidate. Which I plan to, often and repeatedly. Oh, and I can teach a course now if I want.

The proposal defence is one of several milestones on the road to doctoriality. It consists of presenting an outline of what you intend to do for your thesis and what you’ve done to date, and then a fairly gruelling defence session where you stand in front of a committee of professors who take turns interrogating you. Possibly for hours (mine went for about 90 minutes). It’s not exactly adversarial, but it’s not what you’d call “friendly”, either. Though for me it was ultimately just exhausting.

also known as "NIPS", originally uploaded by Mister Wind-Up Bird.

Paper writing is pretty stressful. Code crashes, mistakes are found in equations, and you never have quite the experimental results you want. As deadlines approach, the pressure builds and sleep is abandoned. But I kind of welcome them. And not just for the rare flash of “hey, I’m actually doing science!” satisfaction. A good, hard deadline adds some much-needed structure and discipline to the grad student lifestyle. It forces you to stop tinkering with code and equations and write that shit down. And having just submitted a paper is a great excuse to drink heavily and spend a few days watching movies and playing video games before going back to work.

The paper I submitted yesterday was for the rather grand-sounding Advances in Neural Information Processing Systems. Or: “NIPS”. (For years now, I’ve been thinking about making various off-colour “I heart NIPS” T-shirts, and I swear, one day I’ll do it.) This is my fourth paper since February — the fifth if you count my PhD thesis proposal. I’m kind of looking forward to not writing another one for a while.

This screenshot is my work environment a little before the paper deadline. If you click on that there image, you can see the Flickr version, where I explain what all those windows are doing (more or less: it’s an anonymous-review conference, so I can’t say anything identifying about the research).

On a geeky note (well, even more geeky), I did all the coding, experiments and figures for this paper in Python using Pylab and SciPy, rather than MATLAB. After years of cursing MATLAB (generally by muttering “god, I fucking hate MATLAB” every time it eats up all the available memory and then crashes) and threatening to switch to something — anything — else, I decided I had to either put up or shut up. And shutting up is not my way. Python is not a perfect replacement, but I’m quite happy to report that it worked very, very well and I expect to do most of my thesis work with Python.