Child of a Metal God

A Sunday involving much public transit

livejournal4@cyphertext.net — Tue, 13 May 2008 05:35:30 GMT

Wake up
#22
Seabus
Brunch *omnomnom*
Seabus
#19
Stanley Park
#19
#15
Canadian Tire
#15
#41
Home again
#41
#16
Wrong place argh
#9
Sushi *omnomnom*
#10
#41
Sleep

livejournal4@cyphertext.net — Fri, 09 May 2008 22:31:37 GMT

Holy crap. So, I'm running intro programming labs again this term. Which is tiring, but a lot more fun and interesting than beta-testing assignments again. Today was the first lab with real work in it, and I quickly fell back into my extremely open and encouraging TA persona. Extroverted, willing to expound on random topics, running back and forth with somewhat exaggerated physical movements. I just realized where this all came from.

I'm acting like the Doctor. That completely blows my mind.

my new life

livejournal4@cyphertext.net — Mon, 05 May 2008 21:05:02 GMT

Today I set up at my new desk in the Collaborative Robotics Lab, making me officially part of the Laboratory for Computational Intelligence. This is good for two reason:
1) It is the culmination of a goal I first set on back in high school. Matthew Dockrey, AI Researcher!
2) I now have access to some comfy and fairly private couches for crashing on if I'm stuck on campus.

livejournal4@cyphertext.net — Thu, 01 May 2008 07:44:01 GMT

Also, pardon the video linkage, but this Discovery Channel promo makes me happily weepy. If people can make money promoting a worldview like that, maybe we aren't completely screwed.

livejournal4@cyphertext.net — Thu, 01 May 2008 06:56:13 GMT

Tonight I flew down to Olympia with neuro42 to have dinner with corivax and caladri.

It was a lot of fun! Despite having so many pilot friends I've never managed to go flying with them before. It was a pretty nice day, and the Sound really is gorgeous from the air. I'm happy to report that my map-reading skills haven't atrophied since last summer. (And wow, doing it from a plane is so easy it's almost like cheating. Everything looks just like it does on the map! Much easier than comparing the angles between headlands from a canoe, creeping along at 5 kph.) I got some stick time, which was great, but humbling. That is a skill that will take me a while to learn, when I finally get around to getting my license.

So, in conclusion, whee flying!

Hello, Seattle!

Captain neuro42

Our state capitol building, seen from about 2,000 feet. It was the last impressive state capitol built, before the architectural depths of the 20th century really hit.

Our, uh, mighty steed. Long may it fly!

livejournal4@cyphertext.net — Mon, 28 Apr 2008 20:58:17 GMT

For the record: doing (Canadian) taxes while deathly ill is not fun. But at least my immune system waited until this week to collapse beneath me.

CNN WTF

livejournal4@cyphertext.net — Mon, 28 Apr 2008 06:00:09 GMT

This is the current top story on CNN:

They had a halfway normal pic of Wright up earlier. I guess it didn't convey 'scary black preacher' well enough. Have they stopped even pretending to be impartial? They might as well photoshop in a glowing force-ghost of Malcom X at this point.

whiiiiiiirrrrrrrrrr

livejournal4@cyphertext.net — Fri, 25 Apr 2008 06:50:00 GMT

With 23 minutes to spare, I am done.

A week ago I was dreading this writeup, because I had nothing. Even this morning I was thinking I only had the bare minimum of a real project. As I wrote it, I came to realize it was actually quite interesting. (Though that might just be the Stockholm Syndrome talking.) I still have no idea what kind of marks I'll get, but I'm happy with what I did. I might even have to continue working on this for fun.

I've ended up rather proud of all three of my projects. Machine learning because it went from absolute crap to a simply gorgeous and endearingly quirky paper with real results in just 5 days. Image because the system really truly works and will probably get used in the competition this summer. Even sensorimotor, which I really had been thinking of as nothing more than hardware lols -- I'm really glad I forced myself to go back into the lab yesterday and do another 16 data collection runs, and profile the slow pan irregularities. The paper really benefited from it.

I can sleep again! I can read fiction! I can... start worrying about finding a thesis. :)

Dingdingding

livejournal4@cyphertext.net — Thu, 24 Apr 2008 22:44:33 GMT

Rongorongo
The Tale of Genji
Moby Dick

"...what are three things you don't often seen referenced in a Machine Learning paper?"

"Correct for 200mg of caffeine!"

livejournal4@cyphertext.net — Thu, 24 Apr 2008 19:36:54 GMT

This paper is rated NP HARD by the Mathematical Prudes Association of America.

livejournal4@cyphertext.net — Thu, 24 Apr 2008 12:29:54 GMT

~~422 TAing.~~

~~Image Understanding II.~~

~~Sensorimotor Computation.~~

Machine Learning.

The final push begins. I have 18 hours 34 minutes to get one last paper written. See you on the other side, folks.

livejournal4@cyphertext.net — Wed, 23 Apr 2008 10:52:55 GMT

Acknowledgments

The author acknowledges that this paper sucks.

phase change

livejournal4@cyphertext.net — Mon, 21 Apr 2008 00:20:17 GMT

I feel like I've leveled up in science over the last couple of days. Something finally clicked in my brain, and suddenly my initial reaction to wondering how to approach a problem is 'lit search!'. Which is really duh, I realize, but I've never been the most rigorous person. It's one of the reasons I wanted to do grad school, to break some of my hacky, adhocratic habits.

In a way, it almost feels like cheating. I'm doing less work, getting better results, and I have all these extra papers to cite in the writeup. No wonder this is so popular!

livejournal4@cyphertext.net — Sat, 19 Apr 2008 21:14:16 GMT

I just found myself using Lynx, listening to Depeche Mode. It's 1994 again!

In other news, I'm not dead. Yet. Updates to come as events unfold.

1up

livejournal4@cyphertext.net — Fri, 11 Apr 2008 23:00:18 GMT

I have gained a new special ability: immunity from hepatitis B.

Thank you, socialized health care!

Curling

livejournal4@cyphertext.net — Mon, 07 Apr 2008 19:08:20 GMT

Throwing is a lot harder than it looks. By the end of the day I was mostly able to do it without falling down, but I still ended up with some good bruises on my knees the next morning.

Sweeping without hitting the rock is easier than I would have thought. Sweeping and having it do anything is a lot harder. You really have to push.

Aiming is very hard. One of those was mine, but that was the only decent shot I made. (As you can see, Elan wasn't happy.)

It was fun! It has a nice, sedate pace, with lots of strategizing, and it's actually a pretty good (cardio) workout. Brad is talking about putting together a CS team for next year. I might be able to be talked into that.

Bizarre fact for the day

livejournal4@cyphertext.net — Thu, 03 Apr 2008 22:17:06 GMT

If you've ever heard of the Student's t-distribution, this is just crazy.

The derivation of the t-distribution was first published in 1908 by William Sealy Gosset, while he worked at a Guinness Brewery in Dublin. He was prohibited from publishing under his own name, so the paper was written under the pseudonym Student. --Wikipedia

Weird enough, but then I realized -- under a slightly different corporate climate, this means we could have ended up with a Guiness distribution. Reality is messed up.

presentations

livejournal4@cyphertext.net — Wed, 02 Apr 2008 03:30:16 GMT

In case anyone is interested, here is my Image Understanding II project presentation. It won't make for the best reading, though, since I hate presentations where they just read everything off the slides.

I also need to post my latest UDLS presentation (on pre-GPS navigation techniques), but I need to annotate them a bit first for public consumption. As an experiment they are all horrible crayon drawings on construction paper. It went over quite well, but without the narration it's basically meaningless.

Science!

livejournal4@cyphertext.net — Sun, 30 Mar 2008 10:08:51 GMT

I spent the day working on my sensorimotor project, which is coming along surprisingly well.

This is the eye simulation hardware. It's basically just a camera mounted on a motor with a shaft encoder. It also has a accelerometer, but that isn't being used at the moment. The whole thing sits on a lazy susan for easy rotation. It already had code for simulating saccades (jumps from one position to another) along with a good amount of utility functions, but the only person who knew how to get it running went back to Sweden last December. Part of my project writeup will have to be a proper set of docs for the thing.

For the object tracking part of the project, I needed targets to move around in front of the camera. So I made Charles Babbage and Alan Turing puppets. I certainly don't make them dance up and down or talk in squeaky voices. That would be unprofessional.

The system works using SIFT features, which are basically points in an image that always look the same no matter how the image is oriented, how big it is, or (to a lesser extent) how it was lit. This is the Babbage image with all of its features marked.

Features are then found in frames from the camera. They are matched to features from the training image. Note that most aren't matched, but because there were ~750 to work from, we still get plenty of hits. The code then uses these to determine the center of the object and moves the camera in that direction.

Next steps: The camera motion is currently quite naive and unstable if not heavily dampened. I need to implement something like a proper PID control loop for that, but I'm going to talk to the prof first to find out what is the most biologically correct model to use. I also want to try adding an interface to dynamically choose a region to track instead of using static input images. But really, it's all coming together quite well. With another month left, I can safely move this out of the 'omg stress' category.

a vignette

livejournal4@cyphertext.net — Fri, 28 Mar 2008 18:27:20 GMT

It's Friday morning, so I slept in some. Got up, checked LJ and my standard news sources, took a shower. About halfway through, I notice something drifting past the little frosted window. What could that be? I open it a crack -- snow! Several centimeters of snow! Wtf, climate?

An update

livejournal4@cyphertext.net — Wed, 26 Mar 2008 03:05:40 GMT

What am I up to, you may be asking¹? Well, classes are almost over. I have 2 more assignments to do, one huge and one pretty small. But mostly, now that I have a supervisor, I'm focusing on the 3 class projects due at the end of April. No finals this term, which is nice, but that means projects instead.

Image Understanding II: I'm implementing this paper, adding image features from several others, in order to create a generalized image quality assessment for the SRVC project. Basically, we want to screen out cartoons, paintings, illustrations and renderings, while at the same time ranking the remaining images by how good they are for training a classifier.
Status: Pretty good. I have the paper implemented, and I'm starting to add the new features. I have a decent training set downloaded, but it could use some more work.

Sensorimotor Computation: I'm implementing some visual tracking/servoing on the eye simulation hardware. Not entirely sure what, yet. Basically GYRE in one dimension, with a motor instead of compressed air thrusters. And no freefall.
Status: The person who really knew how to run the system left in December, so it took me about 3 weeks to even get it running again. But this weekend I started working on the new code, which so far hasn't been that hard. I need to implement some kind of PID control loop now.

Machine Learning: For reasons that are too arrogant to be written down, I want to find a way to match the states in two or more Markov processes as equivalent, given just the initial probabilities and the transition table. As far as I and the professor know, no one has ever done this before. And since I'm dealing with (in the test data set) 94 states, I'm basically searching a 94! permutation space. And this is by far my weakest area. So, good times.
Status: lol

1: It's okay if you're not actually asking this.

grad lols

livejournal4@cyphertext.net — Tue, 18 Mar 2008 23:17:57 GMT

I have a supervisor! And only 3 days after the (soft) deadline, too.

Also, I am now a department meeting representative for the CS grad student association, as well as coffee house organizer and (now officially) part of the UDLS committee. But at least there was free pizza and sushi at that meeting.

livejournal4@cyphertext.net — Fri, 14 Mar 2008 07:32:06 GMT

I have had 'Sloop John B' stuck in my head for a good three days now. I'm worried something is going to break in there.

Modern life

livejournal4@cyphertext.net — Wed, 12 Mar 2008 19:05:19 GMT

Things I shouldn't have to reboot yet still have needed to recently:

Cell phone -- This one is borderline, because it's a smartphone. I wouldn't mind so much if it obviously froze or bluescreened, but sometimes it just stops getting signal until I reboot. That's pretty lame.

Microwave -- The microwave that came with the apartment is cheap, underpowered, and prone to ignoring input until I power cycle it. Very lame.

Alarm clock -- Recently my clock has taken to ignoring the alarm settings and just going off randomly, creating false negatives AND false positives. Last night, after fighting with insomnia some more, I finally got to sleep at 3:30. And then the alarm started going off, despite being set for 7:30. Snooze worked, but otherwise I couldn't turn it off, since it wasn't on to begin with. I finally had to power cycle it. Lame to a degree beyond my command of the English language.

wooosh

livejournal4@cyphertext.net — Mon, 10 Mar 2008 20:42:07 GMT

Many many years ago, back when USENET wasn't quite a joke yet, I considered myself a Kibologist. Thus, I'm very excited to see that the Japanese ISS lab module Kibo is finally launching tonight! Also, the super-cool new Canadarm manipulator. I've been waiting to see that in operation for a good 10 years now.