[ExI] DeepMind wins Game1 in Go championship Match

spike spike66 at att.net
Sun Mar 13 03:38:15 UTC 2016

Re: [ExI] DeepMind wins Game1 in Go championship Match

>>... If all you want is a system that can play chess or go, that's clearly 
> doable. But these are more simulated intelligence than artificial 
> intelligence. Produce a system that can be taught any game the way a 
> human learns it, and can learn to play it well via playing and 
> studying the game... -Dave

>...So, what you want is a system that, without knowing anything in advance
about Go, can be taught the rules via natural language and vision, and can
then begin playing and improving its play, maybe even incorporating further
knowledge and analysis of the game ...

What I am looking for isn't so much something that meets our fluid
definition of intelligence, but rather a learning algorithm that can
generalize.  I don't care if we argue over whether it is intelligence or a
clever algorithm.

Imagine a chess playing algorithm that can make inferences and improve its
play.  Then you get two copies of it playing against itself, playing,
playing, 24/7, getting better and better.  OK so we do that with chess.  And
we do it with go.  And we do it with Risk.  Then poker, then insurance, then
sales, and why all the thens?  Why not do all of them simultaneously?

We see where this goes.


More information about the extropy-chat mailing list