Fast Ranking Algorithm: Astonishing Paper by Raykar, Duraiswami, and Krishnapuram

The July 08 (Vol. 30, #7) IEEE Transactions on Pattern Analysis and Machine Intelligence has an incredible paper by Raykar, Duraiswami, and Krishnapuram. A Fast Algorithm for Learning a Ranking Function from Large-Scale Data Sets appears to be a game-changer for an incredibly important problem in machine learning. Basically, they …

more ...

Chess Champ Banned for Bluetooth-in-the-Ear During a Tournament

According to InformationWeek, Umakant Sharma, seeded 2nd in a tournament in New Delhi, was caught with a Bluetooth headset stitched into a cap that he wore "pulled down over his ears" during competition. According to the All India Chess Federation, accomplices fed him moves from a chess program. He's been …

more ...

Cooperative Models For Neither Free-Beer Nor Free-Speech Software

The Netflix optimization challenge exemplifies a situation for which there should be a solution, but which I've not seen a good answer. Namely: for-profit but initially ad-hoc cooperation. For instance, let's just say that I made the case that a Pandora-like "Movie Genome Project" was the key to winning …

more ...

Genetic Algorithm For Kernel Tuning

Scott Swigart points to this article that gives a non-technical overview of the use of genetic algorithms to determine the optimal tuning characteristics of one's Linux Kernel. This ought to work: many years ago I wrote a genetic algorithm that tuned the optimization parameters of one's C++ compiler and it …

more ...

Netflix' $1M Bounty / Amazon's Mechanical Turk...

Netflix is offering \$1M to the first person who can achieve 10% better movie recommendations than their current system. Sweet.

I have all sorts of ideas on this. Thank heavens I have copious spare time.

more ...

Fascinating: Language Exposure Must Be Live, Not Recorded

Infants exposed to a Mandarin-speaking adult for less than 5 hours (25-minute sessions over 4 weeks) were "able to distinguish phonetic elements of that language." Very impressive. But infants exposed to a similar amount of speech delivered over DVD could not. Fascinating. General-audience article here (via The Old New Thing …

more ...

Drools (Java-based inference engine) ported to .NET

Drools.NET is a port of the Drools library to .NET. I have a bg, big architectural decision coming up for a client and I am debating about whether to tackle the issue with an inference engine or a scripting language. So I've been looking at Drools pretty closely. It's …

more ...

Compress Wikipedia, Win 20,000 Euros

Brilliant! The Hutter Prize for Lossless Compression (http://www.hutter1.net/prize.htm) takes as its challenge the task of compressing 100MB of Wikipedia text into the pre-competition best of \~17MB. The idea is that a chunk of Wikipedia text that big has characteristics relevant to compression that go beyond …

more ...

Makers of NaturallySpeaking Raising Expectations for Voice Recognition

NaturallySpeaking 9, coming out in August, claims to dramatically reduce the time it takes to model your voice, achieving the best-possible recognition soon after opening the box.

For some people, that best-possible recognition is said to be 99%. Maybe. I've probably gone throught the "voice training" process a dozen or …

more ...