Re:Is deep learning a new kind of programming?
by CuriouslyC on Sunday December 13, 2020 @09:39AM
(#25406833)
Historically the XOR function has been the simple example that many ML algorithms can't handle. Just imagine a higher dimensional XOR with outliers, and you have a pretty good use case for DL with limited predictors.
[ Parent
| on HN ]
Re:Is deep learning a new kind of programming?
by dr_zoidberg on Sunday December 13, 2020 @03:44PM
(#25410062)
Historically, this was solved in the 80's with the multi-layer perceptron, but it seems it still gets repeated.