"One can imagine an agent that is smart about fi..."

https://arbital.com/p/7pz

by Alexei Andreev Jan 31 2017


One can imagine an agent that is smart about finding and training itself on new features. You seed it with one set of features, but over time it replaces that set with much better features fitting the data. To me it even seems possible that something like that could get to AGI level. This is not "self-modification" in the classic sense, so I'm wondering where that falls in this classification scheme.