Re:Perceptrons – the most basic form of a neural network
by hodwik2 on Sunday June 12, 2016 @06:19PM
(#11890601)
There are arguments that you'd need to create a computer capable of doing "infinite" precision real computation to recreate this effect.
[ Parent
| on HN ]
Re:Perceptrons – the most basic form of a neural network
by pessimizer on Sunday June 12, 2016 @06:48PM
(#11890708)
Where do those arguments come from? I always thought that the problem was that it might be an irreducibly parallel process, and would need its own hardware to simulate cheaply.
That does sound interesting; any links or cites?