BISC: Special BISC Seminar, Friday August 25, 11:00am-12:00pm; 373 Soda Hall


Subject: BISC: Special BISC Seminar, Friday August 25, 11:00am-12:00pm; 373 Soda Hall
From: Morteza Anvari (anvari@eecs.berkeley.edu)
Date: Thu Aug 24 2000 - 12:26:27 MET DST


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

Dynamic Prediction and Learning Technology: Application and Theory
 Overview

 Robert J. (Bob) Jannarone
 Founder and Chairman: Netuitive, Inc.
 Special BISC Seminar
 August 25, 2000

 373 Soda Hall
 11:00am-12:00 Noon

 Abstract

 This seminar will introduce a new dynamic prediction and learning (PAL)

 technology and its underlying theory. Each PAL unit automatically
 operates as follows:

  It obtains input measurement values frequently.
  It delivers predicted output values immediately after obtaining input

 values.
  It learns (i.e., updates prediction equations) before obtaining new
 input values.

 The seminar will outline relative advantages of PAL technology and
 alternative technologies based on statistical, neuro-computing, and
 logic-based methods. PAL technology is best applied to monitoring,
 forecasting, and control under rapidly changing conditions, requiring
 short-term (sub-second to 24-hour) prediction and learning cycles.
 Products based on PAL technology have been operating successfully for
 two
 years. Proven applications include electricity price and demand
 forecasting, along with network performance monitoring.

 The seminar will also outline foundations of PAL theory, including the
 following:
  Computing foundations, including a kernel chip architecture capable
of
 predicting and learning when measurements arrive every few nanoseconds.

  Neural network foundations, including a model for predicting and
 learning in two neuron firing cycles.
  Mental processing foundations, including provisions for reinforced
 learning and perceptual imputing.
  Statistical foundations, including provisions for optimizing
 prediction
 accuracy.
  Fuzzy foundations, including provisions for forecasting and imputing
 when measurements have "viability" values between 0 and 1.

 References:

  Concurrent Learning and Performance Information Processing System
 U.S.
 patent #5,835,902.
  Multi-Kernel Neural Network Concurrent Learning, Monitoring, and
 Forecasting System patent pending (all claims allowed by the
 preliminary
 PCT review).
  Continuation in Part on U.S. patent #5,835,902 including
optimization,
 analog, and other extensions patent pending.
  Jannarone, R.J. (1997). Concurrent Learning and Information
 Processing: A Neuro-computing System that Learns during Monitoring,
 Forecasting, and Control. New York: Chapman & Hall.
  Jannarone, R.J. (1997). Locally dependent models: conjunctive item
 response theory. In W.J. van der Linden & R.K. Hambleton, III (eds.)
 Handbook of Modern Item Response Theory, New York: Springer-Verlag.
  Jannarone, R.J. (1993). Concurrent information processing I: an
 applications overview. Applied Computing Review, 1(2), 1-6.
  Jannarone, R.J. Yu, K.F., & Takefuji, Y. (1988). Conjunctoids:
 statistical learning modules for binary events. Neural Networks, 1,
 325-337, 1987.
  Jannarone, R.J. (1986). Conjunctive item response theory kernels.
 Psychometrika, Vol. 51, pp. 449-460.

 =======================
 Contact Information:

 Dr. Jannarone
  Telephone: 678-256-6102
  Facsimile: 678-256-6110
  Internet: BJannarone@Netuitive.com
  WebSite: http://www.Netuitive.com
  Mail: Netuitive, Inc.
        3460 Preston Ridge Rd., Suite 125
         Alpharetta, GA 30005

 ======================================
 Short Bio:

  Ph.D. in Psychology (Psychometrics), University of California at
 Berkeley, 1980
  M.S. in Mathematical Statistics, University of California at
Berkeley,
 1980
  M.S. in Psychology, University of California at Berkeley, 1978
  B.S. in Computer Science, Florida Institute of Technology, 1972

--------------------------------------------------------------------
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 2b25 : Thu Aug 24 2000 - 12:43:57 MET DST