From cube-lovers-errors@oolong.camellia.org  Sun Jun  1 21:39:09 1997
Return-Path: cube-lovers-errors@oolong.camellia.org
Received: from oolong.camellia.org (localhost [127.0.0.1]) by oolong.camellia.org (8.6.12/8.6.12) with SMTP id VAA01501; Sun, 1 Jun 1997 21:39:09 -0400
Precedence: bulk
Errors-To: cube-lovers-errors@oolong.camellia.org
Date: Sun, 1 Jun 97 19:45:11 EDT
Message-Id: <9706012345.AA19309@sun34.aic.nrl.navy.mil>
From: Dan Hoey <Hoey@aic.nrl.navy.mil>
To: SCHMIDTG@iccgcc.cle.ab.com, cube-lovers@ai.mit.edu
Subject: Re: Searching and Metrics in (Korf 1997)

I wrote:

>I've read through Rich Korf's paper now, and I have a few ideas on the
>paper and how the method might be improved...

Greg Schmidt replies:

> ...And so I feel it is necessary to say that any suggestion for improving
> Korf's method, should take this distinction into account since solving
> the cube is only incidental to the emphasis of the research work that
> professor Korf is involved in....

The method I suggested for reusing tables for new heuristics should be
applicable to any group-theoretic puzzle for which there are
symmetries mapping generators to generators.  For instance, the N^2-1
puzzles have the 8-fold symmetry D4, and so could have one set of
tables used for 16 heuristics.

Given the central nature the memory-performance tradeoff plays in the
paper, I imagine this is quite relevant to Rich's research.

Of course, I also discussed a number of other topics in that message,
but I don't think they were the ones you were addressing in your remarks.

Dan