home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.ai:4730 comp.ai.neural-nets:4686 sci.logic:2532
- Path: sparky!uunet!pipex!bnr.co.uk!uknet!info!L.Newnham
- From: L.Newnham@bradford.ac.uk (Leonard Newnham)
- Newsgroups: comp.ai,comp.ai.neural-nets,sci.logic
- Subject: Neural network controlled theorem proving
- Message-ID: <1993Jan4.145553.16320@bradford.ac.uk>
- Date: 4 Jan 93 14:55:53 GMT
- Organization: University of Bradford, UK
- Lines: 21
- Originator: 90908502@muser
- Nntp-Posting-Host: muser
-
- I am doing research into temporal logic theorem proving. I am thinking
- of using a neural network as a method of learning useful heuristics
- to control the direction of the search. The basic idea is that the
- neural network learns to recognise patterns in the theorem to be
- proved that enable the theorem prover to proceed in an efficient manner
- towards a proof.
-
- I am familiar with the 2 or 3 papers Ertel and Suttner have published
- in this area but have, so far, not been able to find any other research
- along similar lines.
-
- I would be most grateful if anyone could point me in the direction of
- any other research in this area.
-
- Please email and I will post a summary.
-
-
- --
-
- Leonard e-mail: L.Newnham@bradford.ac.uk
-