Bayes-learning-thought etc

From: Martin Lefley (mlefley@bournemouth.ac.uk)
Date: Thu Jan 18 2001 - 18:22:39 MET

  • Next message: Prof. Dr. Siegfried Gottwald: "BOOK ANNOUNCEMENT"

    Hello everyone,

    There is a lot of debate on various issues I'd like to contribute a few
    points that might reduce misconception, verbosity and argument.

    1. Bayesian reasoning is represented by formulae that could be represented
    by other mechanisms. A detailed ANN, FLS or even formulae learning system
    (such as GP) could approximate any Bayesian system.

    2. Probably the cleverest thing about humans is their ability to
    continuously learn. Couple this with enormous, self directed experiences and
    you get a system that will eventually always get it right, it doesn't really
    matter what the underlying paradigm is, just that there is enough
    flexibility and a evaluation function that includes the ability to
    accurately generalise to new situations.

    3. Quantum effects aside (ouch) any system can be modelled by a binary
    logic. Bayesian systems run on binary computers. Nerve cells, at any time
    are firing or not firing. It is important to realise that incredibly complex
    phenomena emerge from very simple functions. The brain is binary but this is
    not understanding or reduction.

    4. Paradigms are probably important only because computers are so
    under-powered (c.f. the brain) so we take short cuts called "models". Most
    likely, one day, super powered parallel computers will out-perform human
    reasoning.

    5. Though fuzzy logic has well defined sub-paradigms the umbrella term
    meaning "many valued logic" can have (m)any forms and hence both includes
    (and is included in wider interpretations of) Bayesian, ANN, GA, MVA, swarm
    intelligence and all other reasoning models!

    Comments welcome, keep them short!
    Martin

    ############################################################################
    This message was posted through the fuzzy mailing list.
    (1) To subscribe to this mailing list, send a message body of
    "SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
    (2) To unsubscribe from this mailing list, send a message body of
    "UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
    to listproc@dbai.tuwien.ac.at
    (3) To reach the human who maintains the list, send mail to
    fuzzy-owner@dbai.tuwien.ac.at
    (4) WWW access and other information on Fuzzy Sets and Logic see
    http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
    (5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html



    This archive was generated by hypermail 2b30 : Thu Jan 18 2001 - 18:37:35 MET