[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

Very impressive result from DeepMind team - Xavier's take?

Posted By: Maik Stiebler
Date: Friday, 22 December 2017, at 4:02 p.m.

In Response To: Very impressive result from DeepMind team - Xavier's take? (Bob Koca)

Yes, that's what the famous Robertie complexity measure does. I certainly had nothing like that in mind, but neither can I contribute anything constructive to the question of how to think properly about complexity with regards to backgammon. I just wrote 'complexity' because I liked the sound of it. What I wanted to express is just that it stretches my credulity that a neural net that can master Chess or Go can't master backgammon including snakes and the other hard stuff. But that statement is really based on nothing and I'm aware that my use of 'master' is vague. I'm especially unsure about the way tactical lookahead and static evaluation interact to produce mastery. That might very well be completely different in all three domains, which complicates the issue. Maybe chess is only difficult/complex for humans and, by extension, neural nets because of the tactics?!

Messages In This Thread

 

Post Response
Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:
 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.