2 ms·
At the lowest level of every computer there's only zeroes and ones. Still you don't try to explain a neural network in terms of zeroes and ones. Because the eme
by Joe-Z 8y ago
At the lowest level of every computer there's only zeroes and ones. Still you don't try to explain a neural network in terms of zeroes and ones. Because the emergent behaviour is much more complex than that.
- lostmsu 8y agoNo, that's actually exactly what you do.
- Joe-Z 8y agoI think we may be talking about different things here. So in the interest of clearing things up: I was using the example of a neural network as "the closest thing" (arguably still very far off) to the human mind we have in computing. If I'm explaining a newly developed software to one of my colleagues I'm talking in terms of services, requests and protocols. Even trying to explain it on the level of an assembly language would be a nightmare. Just think about the millenia it would take me to fully articulate all states the CPU goes through when executing even a single function. And even then all you'd have is an endless list of zeroes and ones, but no explanation as to the meaning of it.