
Genetic Algorithms Digest   Tuesday, 15 January 1991   Volume 5 : Issue 1

 - Send submissions to GA-List@AIC.NRL.NAVY.MIL
 - Send administrative requests to GA-List-Request@AIC.NRL.NAVY.MIL

Today's Topics:
	- Various information from the Moderator
	- De Jongs 5 Functions
	- Re: GA-List v4n27, MIMD implementation
	- GA question: In what instances ... do schemas apply...?
	- Technical Reports (Goldberg)
	- GA and Robotics book info.
	- ICGA 91 - Call for Papers

******************************************************************************

CALENDAR OF GA-RELATED ACTIVITIES: (with GA-List issue reference)

4th Intl. Conference on Genetic Algorithms (v4n17)          Jul 14-17, 1991
          (Deadline for submissions is Feb 1)
AAAI91 - Deadline for submissions is Jan 30, 1991
Machine Learning Workshop - Deadline for submissions is March 1, 1991

(Send announcements of other activities to GA-List@aic.nrl.navy.mil)

******************************************************************************
------------------------------------------------

Date: Tuesday, 15 January 1991
From: Alan C. Schultz (GA-List moderator)
Subject: Various Information

	1)  OK.  I promised more information on Dave Davis' new GA
	book.  Here is what I've got so far:

	Author:    Lawrence (Dave) Davis
	Title:     The GA Handbook
	Publisher: Van Nostrand Reinhold

	2)  The ICGA 91 deadline is coming up on Feb 1.  Therefore, at
	the end of this issue is a reprint of the call for papers for the
	GA conference.

	3)  We need submissions to GA-List if you want it to remain
        active.  The recent lack of issues is due to a lack of submissions.


	Alan C. Schultz

--------------------------------------------------

Date: 11 Dec 90 13:04 +0100
From: Nikolaus Almassy <almassy@ifi.unizh.ch>
Subject: De Jongs 5 Functions

   I am looking for the mathematical definition of the five De Jong test
   functions F1 - F5 defined in K. A. DeJong: "An Analysis of the
   Behavior of a Class of Genetic Adaptive Systems", PhD Diss. University
   of Michigan, 1975.
   I just have the Goldberg book, and there is F2 improperly defined 
   and for F5 there are missing 50 constants

   If anybody has the definition please send me an E-mail.

   Thanks in advance,
   -Nick.

   Nikolaus Almassy
   Department for Informatik         |       Tel + 41 - 1 - 257 43 15
   University Zurich-Irchel          |       Fax + 41 - 1 - 363 00 35
   Winterthurerstrasse 190           |       almassy@ifi.unizh.ch
   CH - 8057 Zurich, Switzerland

---------------------------------------

Date: Mon, 17 Dec 90 19:37 MET
From: DORIGO%IPMEL1.POLIMI.IT@ICNUCEVM.CNUCE.CNR.IT
Subject: Re: GA-List v4n27, MIMD implementation

    Regarding the Ballinger's mail on MIMD implementation.  Here at
    Politecnico di Milano we have developed a MIMD implementation of a
    classifier system on a 18-transputers system (its name is ALECSYS).
    In our implementation it is possible to configure the system as
    follows: Say n is the number of transputers. You can have a single
    CS running on n transputers, n CSs each one running on a single
    transputer or k CSs each one running on h transputers with h times
    k equal n.  We are using our system to test different learning
    strategies and to develop a model of cooperation between different
    contemporaneously active CSs.  In our first experiments we have a
    robot that develops two simple behaviours (it learns to follow a
    light and to avoid heat) and then learns how to coordinate them.
    Anyone interested in it can send me a tech. rep. request.

    A Bootstrapping Approach to
    Robot Intelligence:
    First Results

    Marco Dorigo
    Politecnico di Milano
    MP-AI Project
    Dipartimento di Elettronica
    Via Ponzio 34/5
    20133 Milano
    Italy
    dorigo@ipmel1.polimi.it

    Uwe Schnepf
    Expert Systems Research Group
    German National Research Center
    for Computer Science (GMD)
    P.O.Box 1240
    5205 Sankt Augustin 1
    Germany
    usc@gmdzi.uucp

    Marco Dorigo

------------------------------------------

Date: Wed, 2 Jan 91 10:45:16 CST
From: mcdchg!motcid!marble!ahlenius@uunet.UU.NET (Mark Ahlenius)
Subject: GA question: In what instances ... do schemas apply...?

    Hi,
     I have a basic question on GA's:

    In what instances of GA's do schemas apply?
    Just to classifier type systems or do they apply to 
    things like optimizations such as Goldberg's SGA example?

	    thanks

	    Mark

---------------------------------------------

Date: Thu, 03 Jan 91 13:39:04 CST
From: Dave Goldberg <GOLDBERG@UIUCVMD>
Subject: Technical Reports

    Two technical reports are available from the Illinois Genetic
    Algorithms Laboratory.  Real-coded genetic algorithms, virtual
    alphabets, and blocking (IlliGAL Report No. 90001), describes a
    mechanism of convergence consistent with the schema theorem for
    real-coded GAs (RCGAs) and Evolutionsstrategien.  The paper
    suggests that selection chooses virtual alphabets dimension by
    dimension during the early stages of the GA; these alphabets are
    ultimately searched through the combined action of selection and
    recombination.  The paper uses this theory to predict the existence
    of problems that are blocked from effective RCGA search, because
    the global optimum is surrounded by local optima in a particular
    way.

    Construction of high-order deceptive functions using low-order
    Walsh coefficients (IlliGAL Report No.90002) constructs partially
    and fully deceptive functions using low-order Walsh coefficients.
    This result debunks the folktheorem floating around that high-order
    deception implies high-order non-zero Walsh coefficients.  The
    construction leads to a more complete theory of functions of
    unitation (functions whose value depends only on the number of ones
    and zeroes in the argument).  These ideas are used to help explain
    some of Tanese's puzzling empirical results with parallel GAs
    applied to partial Walsh sums of bounded order.

    To receive copies of either or both of these reports please send a
    request to goldberg@vmd.cso.uiuc.edu.  Include your postal address,
    as these reports are available in hard copy only.

    Dave Goldberg
    Dept. of General Engineering
    University of Illinois at Urbana-Champaign
    117 Transportation Bldg
    104 S. Mathews Avenue
    Urbana, IL 61801
    (217)333-0897

--------------------------------------------

Date: Wed, 12 Dec 90 16:15:10 -0200
From: yuval%wisdom.weizmann.ac.il@CUNYVM.CUNY.EDU (Yuval Davidor)
Subject: GA and Robotics book info.

    Hi Alan,

    I have been receiving several quarries about my book and its
    publication date, so I thought including the below info. would help
    those that are interested.

    The publication date was originally scheduled to August, but only
    early Dec. reached the printers, so it will be available early Jan.
    First copies will be sent to back-log orders and probably after few
    weeks in the shops.

    Thanks and happy new year to all GAlists,

    yuval

    ========

    Series in Robotics and Automated Systems - Vol. 1

    GENETIC ALGORITHMS AND ROBOTICS:
    A Heuristic Strategy for Optimization

    Yuval Davidor, The Weizmann Institute of Science, Israel.


    The material covered in this book is interdisciplinary in nature.
    It combines topics from population genetics, control theory,
    manufacturing technology, and aspects of computer science.  It does
    this by applying a genetic algorithm (GA), a heuristic
    probabilistic search procedure, to the domain of robotics.  Rather
    than reproduce good work already available in the literature on the
    essentials of the above subjects, this book intends to give a
    detailed description of how to apply a GA to real world problems as
    it has not thus far received proper treatment.  I have chosen as a
    model the generation of robot trajectories.  The choice of the
    specific problem was motivated by the fact that trajectory
    generation applies to many process control issues in manufacturing.
    By applying a GA to a real world problem, such as the generation
    and optimization of robot trajectories, considerable insight is
    acquired into the workings of GAs and into the aspects of the human
    interface with complex domains.  It is with this intention and in
    this spirit that this book is presented - to attempt to provide a
    clear introduction to the workings of GAs in the context of
    optimization of large, complex and redundant systems.  It is only
    as a means of introducing and treating the aspects mentioned above
    that this book tries to provide the basic tools for acquiring an
    intuitive understanding as to what GAs are, how they work and when
    it might be rewarding to use them.

	The book is divided into two parts.  Part I describes GAs and
    constitutes an illustrative introduction to the art and practice of
    these special search procedures.  It also attempts to provide an
    answer as to what in genetic algorithms brings computer procedures
    and nature so intimately together.  Part II gives a detailed
    description of an application of a GA model in the generation and
    optimization domain of robot trajectories.  There are three
    chapters at the end of Part II which are devoted to some general
    aspects of learning and adaptation, Lamarckism, Epistasis, and
    emergent global behaviour of local operators.  The subjects of
    these three chapters emerged from the GA model and the considerable
    similarity between the model and natural phenomena.

    Readership: Computer scientists, engineers and applied mathematicians.


    BOOK ORDER FORM:

    Send your order to:

    WORLD SCIENTIFIC PUBLISHING CO. PTE. LTD (Europe Office)
    73 Lynton Mead, Totteridge,
    London N20 8DH, England.
    Telefax: (081) 4463356
    Tel: (081) 4462461

    WORLD SCIENTIFIC PUBLISHING CO. PTE. LTD (US Office)
    687 Hartwell Street, Teaneck, NJ 07666-5309, USA.
    Toll-free: 1-800-227-7562
    Telefax: (201) 837-8859
    Tel: (201) 837-8858

    WORLD SCIENTIFIC PUBLISHING CO. PTE. LTD (Far East Office)
    Farrer Road, POB 128, Singapore 9128.
    Telefax: (65) 3825919
    Tel: (65) 3825663


    Please send:

    Qty: __________ Hard cover @US$28

    Total:US$ ____________

    Method of Payment

    [] Cheque/Bankdraft enclosed for US$ ___________________

    [] Charge my Credit Card:

    [] VISA      []MC      []AmEx      []Diners Club

    Card No.: ______________________________________________

    Expiry Date: _______________ Signature: ________________

    [] Bill my company/institution: ________________________

    Name: __________________________________________________
		      (please print)

    Organization: __________________________________________

    Address: _______________________________________________

    ________________________________________________________

    City: _________________ State: ___________ Zip: ________

    Country: _______________________________________________

    [] Please include my name in your mailing list.

    My field of interest is: _______________________________


    * Prices are subject to change without notice.

------------------------------------------------

Date: Tuesday, 15 January 1991
From: Alan C. Schultz (GA-List moderator)
Subject: ICGA 91 Call for Papers



			      Call for Papers


				 ICGA-91

		    The Fourth International Conference on
 			     Genetic Algorithms



  The Fourth International Conference on Genetic Algorithms (ICGA-91),
  will be held on July 13-16, 1991 at the University of California - San Diego
  in La Jolla, CA.  This meeting brings together an international community
  from academia, government, and industry interested in algorithms suggested
  by the evolutionary process of natural selection.  Topics of particular
  interest include: genetic algorithms and classifier systems,
  machine learning and optimization using these systems, and their relations
  to other learning paradigms (e.g., connectionist networks). Papers
  discussing how genetic algorithms and classifier systems are related to
  biological modeling issues (e.g., evolution of nervous systems, computational
  ethology, artificial life) are encouraged.

  Papers describing significant, unpublished research in this area are 
  solicited.  Authors must submit four (4) complete copies of their paper,
  postmarked by February 1, 1991,  to the Program Co-Chair:

      Dr. Richard K. Belew
      Computer Science & Engr. Dept. (C-014)
      Univ. California - San Diego
      La Jolla, CA   92093

  Electronic submissions (LaTeX source only) can be mailed to rik@cs.ucsd.edu.
  Papers should be no longer than 10 pages, single spaced, and printed using
  12 pt. type. All papers will be subject to peer review. Evaluation criteria
  include the significance of results, originality, and the clarity and quality
  of the presentation.

  Important Dates:

	  February 1, 1991:	Submissions must be postmarked
	  March 22, 1991:		Notification to authors mailed
	  May 6, 1991:		Revised, final camera-ready paper due
	  July 13-16, 1991:	Conference dates


  ICGA-91 Conference Committee:

	  Conference Co-Chairs:	Kenneth A. De Jong, George Mason University
				  J. David Schaffer, Philips Labs

	  Vice Chair and Publicity: David E. Goldberg, 
				    Univ. of Illinois at Urbana-Champaign

	  Program Co-Chairs:	Richard K. Belew, UCSD
				  Lashon B. Booker, MITRE

	  Financial Chair:	Gil Syswerda, BBN

	  Local Arrangements:	Richard K. Belew, UCSD

--------------------------------
End of Genetic Algorithms Digest
********************************

