Congratulations on your new understanding for the computer Go problem. After many more new understandings like this, you will be able to reach the level of understanding many people on this list have about the problem.
Daniel Liu
In a message dated 2/18/05 4:28:15 PM Pacific Standard Time, frank@xxxxxxxxxxxxxxxxx writes:
Evaluation function & pruning is a pretty inefficient way of getting
somewhere. It's like a guy lost in the jungle without a compass or GPS. He
is trying to find an open spot in the jungle a mile away by looking at the
trees around him. When the trees look slightly "open-spottish-like", he
takes that direction.
Well that is silly. He'd better had taken a bearing beforehand and use a
compass!
Pretty trivial if you ask me.
I don't understand why people still want to use search in comp. Go.
I think after decades of trying it's pretty obvious that it doesn't work.
(patterns won't work either, they help search but search doesn't work :-)
The intention is to provoke interesting brainstorming on alternatives to
search! (I mean radical alternatives that are completely novel, so no
patterns, no pruning, no evaluation functions, no domain-specific knowledge
and no tree search of any kind, neither neural networks, fuzzy logic or
genetic algorithms).
I have an idea but it needs some more brainstorming.
My idea, I already explained it in the first paragraph. I am naked before
you and hope you won't betray my trust :)
Thanks in advance for your replies.
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/