Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.mathworks.com!newsfeed.internetmci.com!swrinde!sgigate.sgi.com!enews.sgi.com!news.uoregon.edu!arclight.uoregon.edu!dispatch.news.demon.net!demon!sunsite.doc.ic.ac.uk!warwick!niss!bath.ac.uk!uwe-bristol!banana!g-buckle
From: g-buckle@csm.uwe.ac.uk (G Buckley)
Subject: reinforcement learning neural network
Message-ID: <1996May2.125603.496@pat.uwe.ac.uk>
Keywords: reinforcement learning, neural networks, TD-learning, games, American Football
Sender: usenet@pat.uwe.ac.uk (uwe nntp usenet poster)
Nntp-Posting-Host: usenet@pat.uwe.ac.uk (uwe nntp usenet poster)
Organization: University of the West of England.
X-Newsreader: xrn 7.04-beta-2
Date: Thu, 2 May 1996 12:56:03 GMT
Lines: 23


Hi all,

I'm researching neural network performance in non-deterministic environments,
in particular I've implemented an American Football Coach that calls plays
based on its situation (down, distance, etc..) using reinforcement learning
and limited Temporal Difference learning.  I've been unable to find any
other research into this area, is anyone out there looking into it or know 
someone who is?

The research is soon to be out as a technical paper; 
"Arena Ball: Reinforcement Learning in a non-deterministic environment" 
Gary Buckley, University of the West of England, July 1996, should anyone 
want a gander at a later date.  If anyone has any knowledge of similar 
work I'd be gratefull to hear about it.

Cheers,

Gary Buckley
University of the West of England (UWE)

g-buckle@csd.uwe.ac.uk

