[comp.ai.neural-nets] NN and games

LAZBM@CUNYVM (laszlo grunfeld) (03/15/90)

  Have neural nets been trained to play games or learned evaluation functions?
Any papers written on the subject?

Relay-Version: Version 1.7 PSU-NETNEWS 5/20/88; site MAINE.BITNET
Posting-Version: Version 1.7 PSU-NETNEWS 5/20/88; site MAINE.BITNET
Path: cunyvm!maine.bitnet!michael
From: MICHAEL@MAINE (Michael Johnson)
Newsgroups: rec.food.cooking
Subject: Re: Badmouthing electric woks (was Re: My Chinese food is too greasy. Help!)
Message-ID: <MICHAEL.900315101209@maine.maine.EDU>
Date: Thu, 15 Mar 90 10:12:09 EST
Organization: University of Maine System
References: <1990Mar12.234352.9201@cs.cmu.edu> <23047@mimsy.umd.edu> <MRD.90Mar14132930@image.clarkson.edu>
 <1990Mar14.232107.10345@ultra.com>
TO: NETNEWS@MAINE

In article <1990Mar14.232107.10345@ultra.com> ted@ultra.com (Ted Schroeder)
writes:
>Why do people badmouth non-stick, electric woks?  I use one (I have
>an eletric stove and I'm sure I get better heating with it than I would
>over the stove) and I think it works just great.  Nothing sticks to it,
>it heats quickly, cools quickly, all the things you'd want.  Is there
>some mystical magic about carbon steel woks that is so much better
>that I don't know about?

I would never use a non-stick electric wok for two reasons. You have nowhere
near the heat control of a "real" wok on a gas stove, or even on an electric
stove (you can remove the wok from the heat). The other reason is that when
stir-frying I use a steel spatula and ladle. These handle the heat very well
when I am chow-frying and would chew up non-stick coating in no time flat.
Plastic utensils just don't have the heat capacity or the proper "feel" for
genuine stir-frying. I also suspect that an electric wok would not generate
the heat necessary for stir-frying, but that is an un-founded speculation.

Michael Johnson, University of Maine System           michael@maine.maine.edu

"I will choose the path that's clear. I will choose Free Will."
              -- Neil Peart

Relay-Version: Version 1.7 PSU-NETNEWS 5/20/88; site MAINE.BITNET
Posting-Version: Version 1.7 PSU-NETNEWS 5/20/88; site MAINE.BITNET
Path: cunyvm!maine.bitnet!michael
From: MICHAEL@MAINE (Michael Johnson)
Newsgroups: news.groups,rec.food.cooking
Subject: Re: FINAL RESULTS: rec.food.recipes
Message-ID: <MICHAEL.900315101925@maine.maine.EDU>
Date: Thu, 15 Mar 90 10:19:25 EST
Organization: University of Maine System
References: <1990Mar12.160355.25792@mthvax.cs.miami.edu> <476@donk.UUCP> <608@cica.cica.indiana.edu>
 <1990Mar14.180149.22070@mthvax <38657@iuvax.cs.indiana.edu>
TO: NETNEWS@MAINE

In article <38657@iuvax.cs.indiana.edu> dave@cogsci.indiana.edu (David
Chalmers) writes:
>In article <1990Mar14.180149.22070@mthvax.cs.miami.edu>
aem@Mthvax.CS.Miami.Edu
>writes:
>
>>And I am getting mail as well from people who didn't see their yes or no
>>votes.  The numbers have been equal, and would not have affected the outcome.
>
>Posted results: 233 YES, 116 NO.  Now, assume 10 missing YES votes and 10
>missing NO votes ("The numbers have been equal...").  243 YES, 126 NO.
>Now 243 < 2*126, and the group fails.

This may seem to be a ridiculous idea, but I would like to suggest that those
people who voted either way and didn't see their names/addresses in the voting
summary should POST their votes and AEM can then re-do the tally with these
votes being counted.

Clearly there was a problem here and ignoring the problem is not a fair
reaction.

Michael Johnson, University of Maine System           michael@maine.maine.edu

"I will choose the path that's clear. I will choose Free Will."
              -- Neil Peart

Relay-Version: Version 1.7 PSU-NETNEWS 5/20/88; site MAINE.BITNET
Posting-Version: Version 1.7 PSU-NETNEWS 5/20/88; site MAINE.BITNET
Path: cunyvm!maine.bitnet!michael
From: MICHAEL@MAINE (Michael Johnson)
Newsgroups: rec.autos.driving
Subject: Driving in the DC/Virginia area
Message-ID: <MICHAEL.900315095606@maine.maine.EDU>
Date: Thu, 15 Mar 90 09:56:06 EST
Organization: University of Maine System
TO: NETNEWS@MAINE

I may soon be taking a job in Arlington and I am wondering how the driving is
in that area. I know that radar detectors are outlawed in Virginia. So I'm
thinking I should trade my somewhat bulky Radio Shack unit for a nice, new,
small Escort. So, to summarize my questions:

- How rigorously is the anti-radar detector law enforced? If I put the
  detector on my visor and hide the power cord and keep the lights off (on the
  detector) am I likely to have a problem?

- What areas/times of day are worst to drive in? I'm thinking in terms of
  congestion and boneheads here.

- Are there any areas that are more heavily monitored (e.g. radar) than others?

- Has anyone found any effective methods of camoflaging a radar detector when
  it is mounted up high? I like to have the detector at the top of the wind-
  shield because it is more effective there. My car is built so that I can
  easily run the cord down behind the plastic trim on the roof pillar. It's a
  VW Fox for those who might also own one and could make use of that info. You
  just remove the single screw and then pull the trim off. There are a couple
  of spring clips that will probably fall out when you do this, but they are
  easy to put back in place.

Michael Johnson, University of Maine System           michael@maine.maine.edu

"I will choose the path that's clear. I will choose Free Will."
              -- Neil Peart

ahmad@icsib8 (Subutai Ahmad) (03/17/90)

In article <90074.094831LAZBM@CUNYVM.BITNET>, LAZBM@CUNYVM (laszlo grunfeld) writes:
>
>  Have neural nets been trained to play games or learned evaluation functions?
>Any papers written on the subject?
>

Check out the article "A parallel network that learns to play
backgammon", by G. Tesauro and T. Sejnowski. Aritifical Intelligence
Vol 39, #3 1989.

The network went on to win the backgammon competition at a computer
games tournament in London.  For a short description of this see
"Neurogammon wins Computer Olympiad", by G. Tesauro, Neural
Computation, Vol 1, #3, 1989.

--Subutai Ahmad