How does a giant pile of linear algebra not meet that definition?
Further, this is probably running an algorithm on top of an NN. Some kind of tree search.
I get what you’re saying though. You’re trying to draw a distinction between statistical methods and symbolic methods. Someday we will have an algorithm which uses statistical methods that can match human performance on most cognitive tasks, and it won’t look or act like a brain. In some sense that’s disappointing. We can build supersonic jets without fully understanding how birds fly.
Can you define "almost continuous function"? Or explain what you mean by this, and how it is used in the A.I. stuff?
Is that your point?
If so, I've long learned to accept imprecise language as long as the message can be reasonably extracted from it.
At most you can argue that there isn't a useful bounded loss on every possible input, but it turns out that humans don't achieve useful bounded loss on identifying arbitrary sets of pixels as a cat or whatever, either. Most problems NNs are aimed at are qualitative or probabilistic where provable bounds are less useful than Nth-percentile performance on real-world data.
But, to my mind, something of the form "Train a neural network with an architecture generally like [blah], with a training method+data like [bleh], and save the result. Then, when inputs are received, run them through the NN in such-and-such way." would constitute an algorithm.
When a NN is trained, it produces a set of parameters that basically define an algorithm to do inference with: it's a very big one though.
We also call that a NN (the joy of natural language).
