Good reading. Nice introduction that explains when approaching AI development it's difficult to provide a computer with an exact rigid mechanical model of how the world works, and that's not how we think anyway, but rather we store a lot of small inference models we learned of how things work in general, and then we recall these simple rules to make guesses on what will probably happen in each situation, hence the probabilistic approach to cognition.
Wow. Even starting to explain generative algorithm that can predict a sentence completion or estimate flight of a ball to a dunderhead like me is a tall order - but they have started well
Yeah, sorry about that. Parts of the javascript are packaged using Browserify, but other parts are loaded using a too-clever topological order hack I wanted to try.
This should get fixed by a couple weeks from now (I'm a grad student and this isn't my full-time project :/) but just try reloading the page until you don't get errors.
