I feel like we will end up with a hybrid approach. Powerful statistical models woven together with some kind of logic system..
Re:Prolog's Death (2010)
by marcosdumay on Friday May 10, 2019 @10:07PM
(#19884061)
Well, it's hard to understand why, but for some reason everybody isn't expecting this.
The hype goes on waves, "symbolic AI solves everything!", "no, numeric AI solves everything!".
Re:Prolog's Death (2010)
by TelmoMenezes on Saturday May 11, 2019 @01:46PM
(#19887387)
Exactly!
Re:Prolog's Death (2010)
by thanatropism on Saturday May 11, 2019 @12:07PM
(#19886815)
The maturation and increasing usage of practical SMT solvers is a big sign.
I know this sounds counter-intuitive, but if you've been following such trends for a while, you might agree with me.