Newsgroups: comp.ai.neural-nets
Path: cantaloupe.srv.cs.cmu.edu!nntp.club.cc.cmu.edu!goldenapple.srv.cs.cmu.edu!das-news2.harvard.edu!cam-news-feed3.bbnplanet.com!news.inc.net!uwm.edu!newsfeeds.sol.net!news.maxwell.syr.edu!cpk-news-hub1.bbnplanet.com!su-news-hub1.bbnplanet.com!news.bbnplanet.com!news.pbi.net!uunet!in1.uu.net!uucp5.uu.net!sangam!iitb!bhishma.cse!bhishma.cse.iitb.ernet.in!not-for-mail
From: guru@cse.iitb.ernet.in (S Gurumurthy)
Subject: Connectionist inferencing
X-Nntp-Posting-Host: bhishma
Message-ID: <E7vBJx.L2M@bhishma.cse.iitb.ernet.in>
Lines: 19
Sender: news@bhishma.cse.iitb.ernet.in (USENET News)
Organization: Dept. of Computer Sci. & Engg., IIT Bombay, India
X-Newsreader: TIN [UNIX 1.3 950515BETA PL0]
Date: Sun, 30 Mar 1997 18:04:45 GMT

Hi,
        I  am  at  present  doing  a  project  on   Connectionist
inference of predicate calculus as my M.Tech project. I read some
papers in this, Hinton's DCPS, Shastri's SHURITI, Pinkas work and
Sun's DN model. Of these I think Hinton's DCPS captures the power
of Neural Network(in terms  of  fault  tolerance,  matching  with
incomplete  inputs,  ...). But still the network is too big to be
of any practical use. I am thinking of improving on this line.  I
am thinking of a model with the capabilities of DCPS and which is
efficient. Please send me your comments on whether  it  would  be
feasible or whether it is worth doing.

        email address : guru@cse.iitb.ernet.in

Thanks in advance,

Gurumurthy S.
I M.Tech.
guru@cse.iitb.ernet.in
