Newsgroups: comp.ai.games
Path: cantaloupe.srv.cs.cmu.edu!fs7.ece.cmu.edu!hudson.lm.com!news.pop.psu.edu!news.cac.psu.edu!howland.reston.ans.net!Germany.EU.net!news.dfn.de!nntp.fh-augsburg.de!news.uni-augsburg.de!schaefer
From: schaefer@Uni-Augsburg.DE (Peter Schaefer)
Subject: Re: How to teach a computer to bluff?
Message-ID: <D9JyL1.Ju0@Yuri.CC.Uni-Augsburg.DE>
Sender: news@Yuri.CC.Uni-Augsburg.DE (The News Administrator)
Nntp-Posting-Host: aixcip12.math.uni-augsburg.de
Organization: Computer Center, University of Augsburg, Germany
X-Newsreader: TIN [version 1.2 PL1]
References: <3q2018$l2g@news.bu.edu>
Date: Fri, 2 Jun 1995 15:48:36 GMT
Lines: 23

Yan Lee (y-tony@bu.edu) wrote:
We answer:

The easiest way to get an approximation to bluffing
is to keep track of the opponents moves,
and determine the Minimax strategy( or sth close to it(Bayesian??))
by doing things that hurt less often.

Another thought:
Bluff, and if your opponent believes it, stick to the bluff.

What's bluffing ?
"bluffing is to make a move that would be dumb,
 so it HAS to be another move "

Stratego:
There is one Standard Situation in Stratego that can be analysed:

3 pieces, one of them a spy, can push back a single general,
as long as the location of the spy is unknown.
Just rotate the pieces....

Peter
