← Back to context

Comment by eru

3 years ago

Turing machines can also be used as universal function approximators. But I'm not sure it makes sense to put them in the same category as the other two.

I would love to put them in the same category as the other two. In fact I’ve spent quite a lot if time thinking about it / experimenting. Wouldn’t it be great if we could somehow train on data and get a small Turing machine instead of a huge neural network?