BISC: Nov 15, BISC Seminar

From: masoud nikravesh (nikraves@eecs.berkeley.edu)
Date: Sun Nov 11 2001 - 15:24:13 MET

  • Next message: Colette Faucher: "CFP for KES2002 Special Session "Classification in Knowledge-Based"

    *********************************************************************
    Berkeley Initiative in Soft Computing (BISC)
    *********************************************************************

    Optimality Condition of Interacting Agents and Fuzzy Logic

    BISC Seminar

    Victor Korotkich
    Faculty of Informatics and Communication
    Central Queensland University
    Australia

    Nov 15, 2001
    373 Soda Hall
    4:00-5:00pm

    Abstract:

    Interacting agents, which can range from billiard balls in a random
    interaction to organisms that adapt and learn, offer a
    powerful tool to build models. In the talk we consider the following
    question. Is it possible to have a general optimality
    condition specifying when interacting agents as a whole show their best
    performance for a particular problem? This question has
    significant applied and fundamental interest.
    Recent results of extensive computational experiments give strong facts
    to believe that such condition exists and can be
    formulated in terms of a concept of structural complexity. The
    experiments are made with 100-10000 agents for a number of
    classes of combinatorial optimization problems and by using robot soccer
    simulators.
    In the experiments the structural complexity of agents is monotonically
    increased by a control parameter. This allows us to see
    how the agents' performance changes as they solve the same problem for
    different values of the parameter. A remarkable result
    always appears for the problems tested. Namely, the performance of the
    agents unimodally peaks at some point as their
    structural complexity increases. Hence, the structural complexity of a
    problem can be defined.
    In general, the experiments give considerable computational evidences to
    formulate an optimality condition: the agents show
    their best performance for a particular problem when their structural
    complexity equals the structural complexity of the problem.

    Importantly, the control parameter involved in the optimal condition is
    of fuzzy logic character and admits interpretation in terms
    of human perceptions. This gives a new perspective in the developing of
    models based on interacting agents and show the
    relevance of fuzzy logic to this process.

    --------------------------------------------------------------------
    If you ever want to remove yourself from this mailing list,
    you can send mail to <Majordomo@EECS.Berkeley.EDU> with the following
    command in the body of your email message:
    unsubscribe bisc-group
    or from another account,
    unsubscribe bisc-group <your_email_adress>

    ############################################################################
    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 : Sun Nov 11 2001 - 15:45:04 MET