Newsgroups: comp.ai.games
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.sprintlink.net!newsfeed.internetmci.com!chi-news.cic.net!simtel!swidir.switch.ch!scsing.switch.ch!news.dfn.de!nntp.fh-augsburg.de!news.uni-augsburg.de!news
From: Pit <schaefer@malaga.math.uni-augsburg.de>
Subject: Re: The Hare and Hounds Game
Message-ID: <DGnF3u.F0K@Yuri.CC.Uni-Augsburg.DE>
Sender: news@Yuri.CC.Uni-Augsburg.DE (The News Administrator)
Nntp-Posting-Host: jaen.math.uni-augsburg.de
Organization: Computer Center, University of Augsburg, Germany
References:  <45tikv$glt@ordago.uc3m.es>
Date: Wed, 18 Oct 1995 14:33:29 GMT
Lines: 26

JIgnacio@zenon.uc3m.es (Jose Ignacio Giraldez) 
was looking for programs that play the hare and hounds game.

The game is played on half of the squares of a chessboard.
That means the hares can move diagonally, back and forth, but the
hounds can move only forwards diagonally.

Making a move is mandatory.
The goal of the hares is to break through the lines of the hounds.
The hares loose when they cannot move anymore.

The standard setup is four hounds and two hares.
Other setups work as well, though they might change the side that
has to win.


The problem can be solved by using 'brute force' methods of search,
though i think there should exist a simple method of distinguishing
'bad' and 'good' lines of hounds.

Testing your program should be simple:
The (trained) program should NEVER loose with the hounds,
and win once against a (tired) human.

Bye

