
Genetic Algorithms Digest   Monday, July 26, 1993   Volume 7 : Issue 20

 - Send submissions to GA-List@AIC.NRL.NAVY.MIL
 - Send administrative requests to GA-List-Request@AIC.NRL.NAVY.MIL
 - anonymous ftp archive: FTP.AIC.NRL.NAVY.MIL (Info in /pub/galist/FTP)

Today's Topics:
	- thesis available
	- Paper available via ftp
	- Announcing the availability of the following paper via ftp
	- "Crossover Non-linearity Ratios" Paper Available
	- UK discussion group
	- application of GA to PLA-folding (2 messages)
	- optimization of speech recognizers
	- Information wanted regarding reseach positions and funding
	- Advanced call for papers

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

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

COLT93, ACM Conf on Computational Learning Theory, UCSC (v6n34) Jul 26-28, 93
Machine Learning & Knowledge Acq. Workshop (IJCAI), France (v7n1)  Aug 29, 93
IEE/IEEE Workshop on Nat Alg in Signal Processing, Essex (v7n5) Nov 15-16, 93
AI'93 Workshop on Evolutionary Computation, Melbourne, Aust(v7n16) Nov 16, 93
EP94 3rd Ann Conf on Evolutionary Programming, San Diego (v7n7) Feb 24-25, 94
SPIE, Neural & Stoch. Methods in Image & Sig Proc, Orlando(v7n18) Apr 5-8, 94
The IEEE Conference on Evolutionary Computation, Orlando(v7n10) Jun 26-30, 94
SAB94 3rd Intl Conf on Sim of Adaptive Behavior, Brighton(v7n11) Aug 8-12, 94
PPSN-94 Parallel Problem Solving from Nature, Israel (v7n9)      Oct 9-14, 94

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

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

From: "Egbert J.W. Boers" <boers@WI.leidenuniv.nl>
Date: Thu, 1 Jul 1993 15:39:29 +0200
Subject: thesis available

FTP-host: archive.cis.ohio-state.edu
FTP-filename: /pub/neuroprose/boers.biological-metaphors.ps.Z

The file boers.biological-metaphors.ps.Z (104 pages) is now available for
copying from the Neuroprose repository:

Biological metaphors 
and the design of modular 
artificial neural networks

Egbert J.W. Boers, Herman Kuiper
Leiden University
The Netherlands

ABSTRACT: In this thesis, a method is proposed with which good modular
artificial neural network structures can be found automatically using a
computer program. A number of biological metaphors are incorporated in
the method. It will be argued that modular artificial neural networks
have a better performance than their non-modular counterparts. The human
brain can also be seen as a modular neural network, and the proposed
search method is based on the natural process that resulted in the brain:
Genetic algorithms are used to imitate evolution, and L-systems are used
to model the kind of recipes nature uses in biological growth.

A small number of experiments have been done to investigate the
possibilities of the method. Preliminary results show that the method
does find modular networks, and that those networks outperform 'standard'
solutions. The method looks very promising, although the experiments
done were too limited to draw any general conclusions. One drawback is
the large amount of computing time needed to evaluate the quality of a
population member, and therefore in chapter 9 a number of possible
improvements are given on how to increase the speed of the method, as 
well as a number of suggestions on how to continue from here.

Unfortunately, I'm not in the position to distribute paper-copies of this
thesis. Questions and remarks are most welcome.

Egbert Boers
Leiden University
The Netherlands
boers@wi.LeidenUniv.nl

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

From: dasdan@trbilun.bitnet (Ali Dasdan)
Date: Thu, 15 Jul 1993 17:33:35 +0400 (GMT+3:00)
Subject: Paper available via ftp

The following paper is available via public ftp.

       Genetic Synthesis of Unsupervised Learning Algorithms

                  Ali DASDAN and Kemal OFLAZER
     Department of Computer Engineering and Information Science
                      Bilkent University
                 06533 Bilkent, Ankara, TURKEY
                Email : dasdan@bcc.bilkent.edu.tr

                            Abstract

This paper presents new unsupervised learning algorithms that have been
synthesized using a genetic approach.  A set of such learning algorithms
has been compared with the classical Kohonen's Algorithm on the
Self-Organizing Map and has been found to provide a better performance measure.
This study indicates that there exist many unsupervised learning algorithms
that lead to an organization similar to that of Kohonen's Algorithm, and
that genetic algorithms can be used to search for optimal algorithms and
optimal architectures for the unsupervised learning.

To obtain an electronic copy via ftp, please do the following:

     ftp archive.cis.ohio-state.edu
     login: anonymous
     password: <your email address>
     cd /pub/neuroprose
     binary
     get dasdan.gen-unsup.ps.Z
     quit

Then at your system:

     uncompress dasdan.gen-unsup.ps.Z

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

From: volker nissen <VNISSEN@ibm.gwdg.de>
Date:         Fri, 16 Jul 93 11:50:55 MSZ
Subject: Announcing the availability of the following paper via ftp:

Evolutionary Algorithms in Management Science.
An Overview and List of References.

by Volker Nissen

Contents:
   1 Introduction
   2 Evolutionary Algorithms
     2.1 Overview
     2.2 Genetic Algorithms
     2.3 Genetic Programming
     2.4 Evolution Strategies
     2.5 Evolutionary Programming
     2.6 Classifier Systems and Other Hybrid Approaches
   3 Some Technical Remarks
   4 Applications in Management Science and Related Fields
     4.1 Practical Applications in Business
     4.2 Application-oriented Research
     4.3 General Standard Problems
   5 Conclusions
   6 List of References     <=== about 230 references to research in the field
   7 Questionaire <== I continue to update my data. (Add. material appreciated)

The paper is available via anonymous ftp from the following site:
    gwdu03.gwdg.de

Log in as "anonymous" and send e-mail adress as password. Then change to
directory  pub/msdos/reports/wi .

The paper is contained in postscript-format in the file  earef.eps .

Sincere thanks to all colleagues who responded to my first call for references
last year. Hope you like the results.

Best wishes,

Volker Nissen
Interdisziplinaeres Graduiertenkolleg
Universitaet Goettingen
Gosslerstr. 12a
37073 Goettingen
Germany

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

From: Andrew Mason <esc_mason@ccnov1.aukuni.ac.nz>
Date: 22 Jul 1993 19:59:03 GMT+1200
Subject: "Crossover Non-linearity Ratios" Paper Available

The following paper is now available. Please Email requests to
a.mason@auckland.ac.nz enclosing your postal address and a copy will
be mailed to you. 

Title:
 "Crossover Non-linearity Ratios and the Genetic Algorithm: Escaping
 the Blinkers of Schema Processing and Intrinsic Parallelism."

Author:
 Andrew J. Mason
 a.mason@auckland.ac.nz

Report:
 Report No. 535
 Department of Engineering Science
 School of Engineering
 University of Auckland
 Private Bag 92019
 New Zealand
 
Abstract:
 The initial work in the field of Genetic Algorithms (GAs) sought to
 explain and understand their behaviour by considering such factors
 as Schema Processing and Intrinsic Parallelism. These theories
 underlie much of the research into GAs over the past 2 decades. In
 this paper we examine these contributions in detail and demonstrate
 that while they offer an interesting viewpoint of the GA, they
 cannot be used to argue for the GA's effectiveness.  

 Focusing on crossover as the GA's trademark operator, we consider
 effective crossover partition coefficients and use these to develop
 a new theory of crossover non-linearity ratios. Experimental runs
 are used to demonstrate the power of this theory in characterising
 problems suitable for solution by GAs. 

 We believe that the GA's concept of generating new solutions from
 the best elements of existing solutions is a powerful one. However,
 for this to be accepted by the wider body of optimisation
 practitioners, the GA community needs to adopt a more rigorous
 research approach less dependent upon uncertain evolutionary
 parallels. It is hoped that this work will add momentum to such a
 change.

Keywords: 
 Genetic Algorithms, Schema Processing, Intrinsic
 Parallelism, Implicit Parallelism, Crossover, Deception, Crossover
 Non-linearity Ratios, GA-easy, GA-hard, Crossover-easy,
 Crossover-deceptive, Partition Coefficients, Building Blocks,
 Correlation Coefficients, Tanese Functions.

Andrew J. Mason, Dept of Engineering Science, School of Engineering,
University of Auckland, Private Bag 92019, Auckland, New Zealand

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

From: "s.hedges" <cs0she@orac.sunderland-poly.ac.uk>
Date: Fri, 2 Jul 93 10:16:01 BST
Subject: UK discussion group

We have started up a discussion group for UK researchers in genetic algorithms,
genetic programming, artificial life etc.  If you would like to join send
a message to mailbase@uk.ac.mailbase with the following line in the body:

	join evolutionary-computing <first-name> <last-name>

Regards
Steve Hedges

Department of Computing and Information Systems
Sunderland University		Phone: +44 91 515 2774
Priestman Building		Email: cs0she@uk.ac.sunderland-poly.orac
Green Terrace
Sunderland  UK

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

From: DAVID MOLONEY 0039-2-93330505 <MOLOND@CAST10.ST.IT>
Date: Wed, 7 Jul 1993 11:40:34 DST
Subject: application of GA to PLA-folding

I am interested in finding out if any work has been done on the PLA-folding
problem (a topic in VLSI design) using the GA approach.  My own research has
not turned up any work using GA in this area.  Can anybody help me on this?

Thank you,\

- David Moloney

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

From: DAVID MOLONEY 0039-2-93330505 <MOLOND@CAST10.ST.IT>
Date: Thu, 8 Jul 1993 17:23:19 DST
Subject: GA and PLA-folding

I sent a mail about work on PLA-folding using GA, but work on variable-ordering
problems could also be relevant.  Let me explain.  A PLA consists of rows and
columns which can be manipulated in such a way as to reduce the size of the
final implementation (and hence cost), these folds may or may not be
independent of one and other, so the trick is to find the order of folds that
gives you the longest sequence of independent folds.  I would be grateful for
any references/advice you can give me.

Many thanks,

:-) David Moloney

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

From: toet@izf.tno.nl (Lex Toet)
Date: Wed, 14 Jul 93 17:19:50 +0200
Subject: optimization of speech recognizers

I would like to know if anything has been published on the optimization
of speech recognizers.

Lex Toet

Institute for Human Factors
Soesterberg
The Netherlands
toet@izf.tno.nl

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

From: Rajkumar Roy <rajkumar@charles-cross.plymouth.ac.uk>
Date: Wed, 21 Jul 93 13:56:21 BST
Subject: Information wanted regarding reseach positions and funding

Dear netter,

I am a lecturer of Mechanical Engg. at Jadavpur University, calcutta. I came in U.K. to do M.Sc. (Intelligent Systems) as Rotary Foundation Scholar. I have completed the course.I have Post graduate degree in production engg. from Jadavpur univ., I have also three years industrial experience at TELCO.

I am interested to complete my Ph.D. on neural networks or Genetic algorithms application in mechanical engg. problems in U.S. or U.K.

I have got admn. in Penstate Univ. and Tennessee Tech. Univ. in US and Univ. of Strathclyde. But have not got any fund.

Can any body give any kind of suggestions, informations regarding research pos., funding etc. Any help will be highly appreciated. If required I can send CV and recent work abstract to individuals.

Please write to me directly:

Till 28th July'93 : rajkumar@uk.ac.plym.cx

After 28th July'93 :

RAJKUMAR ROY
11/6/1 JHEEL ROAD
NEWLAND
CALCUTTA - 700075
INDIA

I look forward to your kind response.

Thanks,

rajkumar roy.

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

From: "Leigh Tesfatsion" <S1.TES@ISUMVS.IASTATE.EDU>
Date:      Fri,  9 Jul 93 22:02:28 CDT
Subject: Advanced call for papers

I am on the Program Committee for the next North American Summer Meeting
of the Econometric Society, to be held June 24-28, 1994, in Quebec.   My
area  of  responsibility is self-organization and emergent behavior.  If
your research  relates  to  these  topics,  I  hope  you  will  consider
submitting a paper for this conference.

A detailed announcement and call for papers  for  the  conference  as  a
whole  will appear soon in _Econometrica_.  All paper submissions should
be mailed to  the  Program  Committee  Chair:   Professor  M.   Wooders,
Department  of Economics, University of Toronto, 150 St.  George Street,
Toronto, Ontario M5A 1A1,  CANADA.   The  deadline  for  submissions  is
January 25, 1994.

Each submission should include three copies of each of the following:  a
cover page, an abstract,  and  (if  possible)  the  paper  itself.   The
recommended  length  of  abstract  is  100 words.  The cover page should
include:  the name of the author submitting the paper for  presentation;
the names, institutional affiliations, addresses, telephone numbers, and
(if available) the email addresses of all the authors; the paper  title;
a  single  J.E.L.  (_Journal of Economic Literature_) primary field name
and number; and up to three optional field designations  from  among  17
listed  fields.  (For GA-List members, the most relevant of these fields
would seem to be (4) game theory, (16)  information  and  learning,  and
(17) self-organization and emergent behavior.)

Thanks!

Leigh Tesfatsion
Department of Economics
Iowa State University
Ames, IA 50011-1070

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

