computer-go: Insight of a human

[Vincent Diepeveen <diep@xxxxxxxxx>]

At 10:15 PM 8/31/00 +0200, you wrote:
>> >Moreover, such an intelligent program should
>> >be definitely faster than a brute-force one (on the same hardware) - and
>> >thus better!
> 
>> The conclusion sounds weird to me: why would something more selective
>> be faster? For brute force you don't lose speed. You do everything.
>> When being selective
>>   a) you lose speed to the selection
>>   b) smaller branching factor ==> more speed loss (partly result of a)
>
>Well, what I had in mind when saying that is that a brute-force approach
must go both deep and wide in order to play well. A really smart program
that might for example think like a human, won't have to deal with the same
amount of calculations, thus becoming faster. Maybe (notice I said "should
be faster")

There is however one small problem: how do you program in 0s and 1s
the oversight/strategich insight which even an average human has?

You can put in facts like: "if this pattern then be happy", but the
oversight is real hard. Usual the oversight for programs is the nowadays
luck of the summation of patterns, where a human is doing all this way 
smarter.

Any thoughts on this?

So not the search part but the evaluation part?

Suppose you have n patterns which are all independant from each
other. Right now

  eval = sum ( n )

A bit primitive.

>Vlad
>
>
>
>