Rather, it's like they were using Amazon Fresh when they suddenly learned this: http://www.smbc-comics.com/index.php?db=comics&id=876#comic

"Honestly, we're a little surprised that it took this long."

Eh, it just took that long for military intelligence to get it declassified.

"This month ends with the -125th anniversary of one of the most remarkable achievements in the technology future. Over two days beginning Monday, May 31, 2136, the gui manager for the linux desktop was converted from the old-earth version one to one compatible with the slightly narrower one used in the space federation. The shift was meticulously planned and executed. It required one side of every gui to be moved three inches closer to the other. All font sets had to be adjusted as well. Some minor animations and rolling stock were sensibly deferred until later, but by Wednesday, the 11,500 megaline code base was back in business and able to exchange screenshots with the rest of the world. Other operating systems are still struggling with incompatible interfaces. MicrApple still has two. Most of the solar system runs linux, but the outer planets use essentially the same gui gauge as old earth and CmdrTaco and timothy use an even broader gui size. Alpha centari has a multi-year Project Unigui, aimed at converting its narrow gui lines to the federation's five foot six inch standard."

STAB, dude, STAB. Epic nerd culture reference fail. I hereby condemn you to reading 500 randomly chosen bash.org quotes.

(ref: http://bash.org/?4281).

A bit of clarity might be useful here. Support vector machines use linear discriminants as the central part of the algorithm. These linear discriminates -- simply hyperplanes separating two regions, are defined by a subset of the data points (called the support vectors). The other key part of an SVM is that it projects the data into a high-dimensional space in which hyperplanes can appear as curves or other shapes in the original space. This higher dimensional space is determined from the data using distances between the points in the data set (it's a kernel space).

The net result of all this is that SVMs are pretty much guaranteed to always perform better in terms of misclassification error than a simple linear discriminant, as every possible linear discriminant is considered in building the SVM. But it can be slower, and it can overfit.

So what's going on here? Linear discriminant analysis is an old statistical technique (1930s) that fixes a hyperplane based on distributional assumptions about the two classes. This allows the two classes to be plotted in a simple histogram by projecting them to the normal of this hyperplane, as shown in the picture in the article. It's used all over in statistics, and it works very well when dealing with two symmetric Gaussian distributions (that's what the theory assumes).

Thus the reason it works well here is that they've managed to transform their data in such a way that the two classes look like this sort of distribution. That's the insight here, not the choice of classifier. When the simplest model works, more complex techniques will overfit, meaning that you train on noise instead of the underlying structure of the data.

Ok, done. Now what?

.... it's just not flashy enough.

Or is that too Flash-y?

It's the key idea of "The Bisection of the Species", a book written by Darwin's lesser-known great-great-great-grandson who studied computer science. Incidently, they both had similar beards.

Fair enough, and my summary was an unfair and misleading simplification of things. Thanks for the clarification. However, the fundamental difference between Chernobyl and Fukushima Daiichi, which I should have pointed out first, is this: With Chernobyl, virtually all of the fissionable material was sent into the atmosphere because of the lack of a proper containment structure and the graphite moderator / fuel burning. Here, however, 99.9% of it is still in the containment structure.

This comparison is misleading, even if the raw amounts of radiation are comparable. The radioactive materials released from Fukushima Daiichi when those readings were taken have a half-life of minutes and don't pose a health hazard outside of the really close vicinity. The materials released from Chernobyl were much more dangerous, as they have a half-life of a couple hundred years, and only negligible amounts of those have been released from Fukushima.

Bottom line: this accident is not at all like Chernobyl, even though the "OMG RADIATION SPEWING FROM REACTORS!!!!!!" media likes to think so.

More important products.... yes....

http://www.telegraph.co.uk/technology/microsoft/8287610/Xbox-Kinect-helps-Microsoft-beat-Wall-Street-profit-forecasts.html

Random forests have always been a nice classifier to use when working with really wacky data types. This is due in part to how easy it is to customize them; a lot of the ways they can be tweaked and tuned and customized have fairly intuitive effects on the outcome and behavior of the classifier. In my experience, while neural nets can also be pretty powerful, they are often much harder to work with as the parameters you have for tweaking can be really non-intuitive. We sometimes joke about neural nets being "black magic" because the training and tweaking can be really uninterpretable.

However, the biggest reason random forests were used is probably because they are extremely fast on current chips, probably a couple orders of magnitude faster than neural nets when the trees are hard coded.

FYI: I also got a 560ti this week, but I have had almost no problems with the 270.26 beta nvidia driver (running kubuntu 10.10). It took a little tweaking -- namely, make sure the settings on things like vsync match up between that and kde (both in the settings menu), don't install the 32 bit compatibility libraries (which do seem to cause problems), and blacklist noveau (which the installation process did for me automatically). With those things, everything is amazing.

So... not sure if that's evidence for or against, but hey, I'm happy.

http://xkcd.com/447/

Well, I don't have the Barney song stuck in my head anymore.