Stay updated with breaking news from Universal Turing. Get real-time updates on events, politics, business, and more. Visit us for reliable news and exclusive interviews.
I've supervised a lot of great student projects in my nine years at MIT, but my inner nerdy teenager has never been as personally delighted by a project as it is right now. Today, I'm proud to announce that Adam Yedidia, a PhD student at MIT (but an MEng student when he did most of this work),…
In 1969, Marvin Minsky and Seymour Papert published Perceptrons: An Introduction to Computational Geometry. In it, they showed that a single-layer perceptron cannot compute the XOR function. The main argument relies on linear separability: Perceptrons are linear classifiers, which essentially means drawing a line to separate input that would result in 1 versus 0. You can do it in the OR and AND case, but not XOR. Of course, we’re way past that now, neural networks with one hidden layer can solve that problem.
Bird Song of the Day From Ethopia. I thought cranes would have more interesting calls than they do, but they all honk in more or less the same way. * * * #COVID19 At reader request, I’ve added this daily chart from 91-DIVOC. The data is the Johns Hopkins CSSE data. Here is the site. I feel I’m engaging in a macabre form of tape-watching. All the charts are becoming dull — approaching nominal, if you accept the “new normal” of cases, for example. Still whoops. I guess we’ll see if Biden abandoning masks provides sufficient incentive. “Nearly 3 in 5 Unvaccinated Adults Say a Big ...
LESSWRONG 88 ( Originally posted in December 2015: A dialogue between Ashley, a computer scientist who's never heard of i. Unbounded analysis BLAINE: Good evening, Msr. Ashley. ASHLEY: I've heard there's this thing called "Solomonoff's theory of inductive inference". BLAINE: The rumors have spread, then. ASHLEY: Yeah, so, what the heck is that about? BLAINE: Invented in the 1960s by the mathematician Ray Solomonoff, the key idea in Solomonoff induction is to do sequence prediction by using Bayesian updating on a prior composed of a mixture of all computable probability distributi...