Unprovability comes to machine learning

Re:Unprovability comes to machine learning by Rainymood on Wednesday January 09, 2019 @06:19AM (#18863907)

>For me, the interesting conclusion is that there are things that humans are mathematically incapable of knowing.

Define

>mathematically

>incapable

>knowing

These hastily drawn pop-sciency conclusions are part of the problem why ML/AI has garnered so much media attention these days ...

[ Parent | on HN ]

Re:Unprovability comes to machine learning by Frenum on Wednesday January 09, 2019 @09:39PM (#18871070)

I probably should have used the word "implication" instead of "conclusion". My thought process was that there are certain input functions that the learning structures within the human neural network can't discern.

[ Parent | on HN ]