Re: An AI program that doesn't learn

[Patricia Hughes and David Elsdon <babel17@xxxxxxxxxx>]

"P.J.Leonard" wrote:

> > 3. Groups of stones - I'm finding this hard to refine enough to be useful
>
>  I agree it is hard and it is not useful.
>
>  For starters  . .  .  what is a group ?

A group is a collection of co-located strings - and we all have an intuitive
idea of what a group is [though we would not agree necessarily in all cases],
so that is what I aim to recognise in my program. At the moment strings whose
fields of influence intersect by a "certain amount" constitute a group. At the
moment I am tuning "certain amount" so that it captures "my" intuition of what
a group is.

>  I think it useful to be able to ask questions about sets of strings.
>
> e.g. can any string in the set live ?  can all the strings in the set
> live ?
>      can strings connect ?  can strings be cut ?
>
>  You could then investigate the stability of various combinations of
> strings.
>
>  But to think of a set of stones as a concrete group  and keep track of
> it's
> identity as the game progress . . .  Hmmmm ?

I do think it is necessary to keep track of a group's identity. My program has
no notion of the history of the game. It determines its move from the current
position and nothing more. And I am convinced that this is the right way to do
it.

> --
>  Cheers Paul (P.J.Leonard)
>
>          ``Reality is for people who lack imagination''

I think people who always finish their emails with ``Reality is for people who
lack imagination'' lack imagination :-)

Regards

David