
Genetic Algorithms Digest   Tuesday, May 11, 1993   Volume 7 : Issue 12

 - 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:
	- Re: New Release of Genocop (v7n3) - The things people do !
	- GAs on supercomputers
	- digests in simulated annealing and numerical analysis
	- Free Books/Software
	- Innsbruck conference report
	- Evolutionary Robotics
	- ECAL 93 - FINAL PROGRAM
	- IEE Colloquium: GAs in Control Systems Engineering
	- FLAI93-exhibition

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

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

ECAL-93, 2nd European Conference on A-Life, Brussels (v6n31)    May 24-26, 93
CSCS93, 9th Int Conf on control systems & CS, Romania (v7n3)    May 24-27, 93
ANN93, IEE Intl Conf on Artificial Neural Nets, Brighton        May 25-27, 93
ICGA-93, Fifth Intl. Conf. on GAs, Urbana-Champaign (v6n29)     Jul 17-22, 93
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
EP94 3rd Ann Conf on Evolutionary Programming, San Diego (v7n7) Feb 24-25, 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: zbyszek@mosaic.uncc.edu (Zbigniew Michalewicz)
Date: Thu, 15 Apr 93 10:03:54 EDT
Subject: Re: New Release of Genocop (v7n3) - The things people do !

From: "Andy Keane" <andy.keane@engineering.oxford.ac.uk>

>   I was interested to see the benchmark given for the new release of
>   Genocop (version 2.0) in vol 7 no 3 of the digest:-
>
>     This is evident, for example, for the following problem 
>     (Colville function):
>
>       minimize  100(x_2 - x_1^2)^2 + (1 - x_1)^2 + 90(x_4 - x_3^2)^2 +
>		      + (1 - x_3)^2 + 10.1((x_2 - 1)^2 + (x_4 - 1)^2) + 
>		      + 19.8(x_2-1)(x_4 -1),
>     where -10 <= x_i <= 10, i=1,2,3,4; with the global solution (1,1,1,1) and 
>     f(1,1,1,1) = 0.
>
<     The typical solution found by the original Genocop system in 1,000,000 
>     (million) generations is 
>	  (0.983055, 0.966272, 1.016511, 1.033368)
>     with f = 0.001013, whereas a typical solution returned by the 
>     new version in 10,000 generations only (i.e., 1% of the original time), is
>	  (1.000581, 1.001166, 0.999441, 0.998879)
>     with f = 0.0000012.
>
>   I would agree that this is quite tough to solve using GA's. However,
>   using the linear approximation routine APPROX given by J.N. Siddall in
>   his book "Optimal Engineering Design: Principles and Applications",
>   Marcel Dekker, Inc. 1982, New York the correct solution of (1,1,1,1)
>   is obtained with just 150 tries, i.e. ONE generation of 150 members.
>   It seems to me that the GA community would be well advised to be
>   more aware of and deploy other techniques alongside those being
>   developed within the GA world. Some of them are quite good and
>   often much more appropriate to the problem in hand !

It seems that Andy Keane misunderstood my announcement.

In September 1992 the first version of Genocop was released (see v6n31).
For the next five months I worked on the set of operators incorporated
in Genocop. In most cases I left their original names (like "arithmetical
crossover", "non-uniform mutation", etc.), but I rewrote the procedures behind
them. As a result, a new version of Genocop was ready in February this
year (v7n3). The new version is significantly better than the original one;
a Colville function (cited above) served just as an example for which
a significant speedup between the original and next version was achieved
(and it was not any "benchmark" as suggested by Andy Keane).

I did not make any claims that the Genocop is the world record holder
for efficiency on Colville function, which is an example of an
unconstrained optimization problem with differentiable objective
function.  Clearly, there are many other methods which would approach 
the Colville function in much more efficient way. As an example, 
note that it is very easy to develop a "system" to optimize
a LINEAR function of ONE variable, where the variable is bounded in the
range <a,b>. Such a "problem" can be solved immediately without any 
calculations (just one "if" statement would do), and the 
Genocop would still require some time. What does it prove? 

Genocop handles optimization problems where the linearly constrained 
objective function is (possibly) nondifferentiable or discontinuous, with
(possibly) many local optimas. In the same time, it is useful exercise
to test the system on standard test functions with known optimas: it is
why I used the Colville function at some stage. It would be an 
interesting exercise, if Andy Keane reports on comparison of Genocop 
and the linear approximation routine APPROX on nondifferentiable and
discontinuous multimodal functions. His comment that "there exists a function, 
for which some other algorithm performs better than Genocop" is meaningless.

Zbigniew Michalewicz

Mail: Department of Computer Science      E-mail: zbyszek@mosaic.uncc.edu  
      University of North Carolina        Phone:  (704) 547-4873           
      Charlotte, NC 28223                 Fax:    (704) 547-2352           

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

From: "tettamanzi" <tettamanzi@hermes.mc.dsi.unimi.it>
Date: 27 Apr 1993 16:16:10 U
Subject: GAs on supercomputers

Hi,

I am working on GAs at the Computer Science Dept. of the State University
of Milan; we are planning to carry out some experiments on a supercomputer,
but we do not have one in our campus, so we are searching for a remote site
suitable for the purpose.

Since we already have working software written in C++, it would do best
if a C++ compiler were available; moreover it would be nice if someone
out there had already done something similar somewhere and could give us
a pointer.

Another question is: does anybody know of a supercomputing site supplying
a GA support system (I am thinking of something like Whitley's Genitor II)?

Any response will be greatly appreciated.

Andrea Tettamanzi

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

From: amit@cc.iitb.ernet.in
Date: Thu, 22 Apr 93 15:00:34 +0630
Subject:   digests in simulated annealing and numerical analysis

Sometimes back I sent a request via GA-digest for info regarding the digests
available in the different fields of optimization. Unfortunately it failed
to evoke much positive response. It seems we have a long way to go for
a full fleged digests in different spheres of optimization!! Anyway below
are given addresses of a few digests which our community may find interesting.

     Simulated Annealing   : anneal-request@cs.ucla.edu
     Numerical Analysis    : na.help@na-net.ornl.gov

Just send e-mail to them requesting a subscription. Be patient as the
processing is likely to take some time. I sent them a request a month back
and have recd only test messages from them!!

Amit Kumar Sinha            email: <amit@cc.iitb.ernet.in>
Research Scholar
IIT, Bombay, INDIA

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

From: John Scales <jscales@dix.Mines.Colorado.EDU>
Date: Thu, 22 Apr 93 09:57:28 -0600
Subject: Free Books/Software

We have decided to give away some lecture notes and software that
we've  written  over the years.  More is in the works and we hope
that others will join us in making their works available for  all
to  share.   None  of  the authors, the Colorado School of Mines,
New England Research, or anyone else you can  imagine  makes  any
promises or guarantees about anything in these documents/codes.

If you want any of this material, here's what you do.  Ftp to the
Internet  site  hilbert.mines.colorado.edu (138.67.12.63) and log
in as anonymous.  Any password will do.  Dir doesn't work here so
you  need to use ls (or mls if you're coming in from abroad).  CD
to pub and you will see, among other things, the following direc-
tories:

papers
uga

Directory  uga  contains  a  preliminary  release  of  UGA  (uni-
processor genetic algorithm) and associated class libraries.  The
code was developed on Sun Sparcstations  using  AT&T's  C++  com-
piler.  Versions which will compile using the GNU C++ compiler on
various architectures are in preparation.

The lecture notes are in  directory  papers.  Subdirectories  in-
clude:

migration
conjugate_gradient
theoretical_seismology

What you will see in these directories are compressed  postscript
files.  Migration contains lecture notes for a graduate course in
seismic imaging.  It's about 200 pages long and  has  roughly  55
figures.

Conjugate_gradient contains notes for a short course on  the  use
of  sparse  matrix  methods  for  inverse calculations.  About 40
pages with no figures--yet.

Theoretical_seismology contains  lecture  notes  for  a  graduate
course  in  theoretical  seismology.   About 120 pages with a few
figures. We expect that many figures will be added  in  the  next
few months or so, at which time we'll put out a revised version.

So, there you have it.  If you have any books, lecture  notes  or
software  that you would be willing to distribute freely, drop us
a line and we'll be happy  to  include  your  contribution.   And
remember,  you get what you pay for:  please don't send any irate
email asking how you get from equation 8.4.6 to 8.4.7 on page  so
and so of some document.

                         Samizdat Press

John A. Scales                        Martin L. Smith
Center for Wave Phenomena             New England Research
Deptartment of Geophysics             White River Junction
Colorado School of Mines              Vermont, 05001
Golden, Colorado 80401                martin@ner.com
jscales@dix.mines.colorado.edu

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

From: CRReeves <srx014@cck.coventry.ac.uk>
Date: Mon, 26 Apr 93 16:39:02 WET DST
Subject: Innsbruck conference report

The International Conference on Artificial Neural Networks and Genetic 
Algorithms was held in Innsbruck from April 13-16. 

About 100 papers were presented, mostly of a very good standard, about 
half of them devoted to GAs, including a number dealing with applications 
of GAs to neural nets. There were applications in combinatorial 
optimization, marketing, control theory, pattern recognition, robotics 
etc. There were several interesting studies of parallel implementations, 
links with other heuristics (e.g. tabu search), and quite a number on 
various aspects of classifier systems.

One of the recurring themes that emerged from several papers and in 
subsequent discussions was on the need to use GAs (and ANNs as well) 
appropriately - that is, to make sure that the problem domain is a real 
challenge for the GA, and not something where faster traditional methods 
would perform better: an interesting echo of the recent comments in the 
Digest on comparing GAs with tailored heuristics.

The proceedings have been published by Springer-Verlag in Vienna. If 
anyone wants a copy, the conference organisers have some left which 
they are able to sell at a discount - 1600 Austrian schillings instead 
of 2000. At that price, I guess it's mainly for libraries, but there is 
some good stuff in there. For details, contact 

Fr Anita Zuderell, Institut fur Informatik, Universitat Innsbruck, 
Technikerstrasse, A-6020 Innsbruck, Austria.
Phone: 0512-218-5240
Fax: 0512-218-5252

-- 
| Colin Reeves				    |
| Division of Statistics and OR		    |
| School of Mathematical and Information    |
| Sciences             			    |
| Coventry University			    |
| Priory St				    |
| Coventry CV1 5FB			    |
| tel :+44 (0)203 838979		    |
| fax :+44 (0)203 838585		    |
| email: CRReeves@uk.ac.cov.cck		    |
| (alternative email: srx014@cck.cov.ac.uk) |

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

From: A.Fraser@eee.salford.ac.uk
Date: 26 Apr 93 16:43
Subject: Evolutionary Robotics

Hi,

     We are currently producing an overview of work in Evolutionary Robotics
for journal submission. If you are working in the field (or those mentioned
below) a paragraph detailing your work and references to it would guarantee
your entry. If still in the limbo stage before publication we will gladly take
internal documents for references. Also pointers (email or snail addresses) to
those carrying out research would be gratefully recieved

We concieve Evolutionary robotics as a fusion between artificial life and the
bottom-up approach to robotic control architectures. We also believe that while
real world automata is preferable temporal constraints from computational
limitations mean a lot of the work is carried out in simulation, this is
completely valid and will be included.

Please send applications to:

A.Fraser@eee.salford.ac.uk
J.R.Rush@eee.salford.ac.uk


 Apologies:This is being sent out to a number of
lists if you recieve it more than once we are sorry...

Thanks in advance,

     Adam P. Fraser & Jon R. Rush

=============================================================================
|| ||\\   /|| ||\\    ||\\     //\\    || A.P.Fraser,                      ||
|| || \\ / || ||  \\  ||  \\  //       || Snail:PostGraduate Section       ||
|| ||   /  || ||  //  ||  // //   \\\  ||       Elec & Electronic Dept     ||
|| ||      || ||//\   ||//\   \\   //  ||       University Of Salford      ||
|| ||      || ||   \\ ||   \\   \\//   ||       Salford, M5 4WT, England   ||
|| ...Mobile Robots Research Group...  || Email:A.Fraser@eee.salford.ac.uk ||
=============================================================================
||||        When we try to pick anything out by itself,  we find         ||||
||||    it hitched to everything else in the universe.  - John Muir      ||||
=============================================================================

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

From: sgoss@ulb.ac.be (Goss Simon)
Date: Tue, 4 May 93 11:21:57 MET
Subject: ECAL 93 - FINAL PROGRAM

ECAL '93

2nd European Conference on Artificial Life

SELF-ORGANIZATION AND LIFE, FROM SIMPLE RULES TO GLOBAL COMPLEXITY

Brussels, May 24-26th, 1993

Natural and artificial systems governed by simple rules exhibit self-
organisation leading to autonomy, self-adaptation and evolution. While these 
phenomena interest an increasing number of scientists, much remains to be done 
to encourage the cross-fertilisation of ideas and techniques. The aim of this 
conference is to bring together scientists from different fields in the search 
for common rules and algorithms underlying different systems. The following 
themes have been selected :

- Origin of life and molecular evolution
- Patterns and rhythms in chemical and biochemical systems and interacting cells
 (neural network, immune system, morphogenesis).
- Sensory and motor activities in animals and robots.
- Collective intelligence in natural and artificial groups 
- Ecological communities and evolution .
- Ecological computation.
- Epistemology

We are also planning demonstrations of computer programmes, robots and 
physico-chemical reactions, both in vivo and in video.

Invited Speakers

C. Biebricher (Germany), S. Camazine (USA), H. Cruse, P. De Kepper (France), 
W. Fontana, N. Franks (UK), F. Hess (Holland), 
B. Huberman (USA), S. Kauffman (USA), C. Langton (USA), M. Nowak (UK), 
T. Ray (USA), P. Schuster (Germany), M. Tilden (Canada), J. Urbain (Belgium),
F. Varela (France).

Organising committee

J.L. Deneubourg, H. Bersini, S. Goss, G. Nicolis (Universite Libre de Bruxelles)
R. Dagonnier (Universite de Mons-Hainaut).

International Program committee

A. Babloyantz (Belgium), G. Beni (USA), P. Borckmans (Belgium), 
P. Bourgine (France), H. Cruse (Germany), G. Demongeot (France),
G. Dewel (Belgium), P. De Kepper (France), S. Forrest (USA), N. Franks (UK),
T. Fukuda (Japan), B. Goodwin (UK), P. Hogeweg (Holland), M. Kauffman (Belgium),
C. Langton (USA), R. Lefever (Belgium), P. Maes (USA), J.-A. Meyer (France),
T. Ray (USA), P. Schuster (Austria), T. Smithers (Belgium), F. Varela (France),
R. Wehner (Germany). 

Address:

ECAL '93,
Centre for Non-Linear Phenomena and Complex Systems, CP 231,
Universite Libre de Bruxelles, Bld. du Triomphe, 1050 Brussels, Belgium.
Fax : 32-2-6505767; Phone : 32-2-6505776; 32-2-6505796;
EMAIL : sgoss@ulb.ac.be

[Ed's Note:  This message has been shortened due to space constraints.
The full message, which includes the registration form and final program,
is available from the FTP server, ftp.aic.nrl.navy.mil in the
file /pub/galist/information/conferences/ECAL93-FINAL  -- Connie]

Simon Goss

Unit of Behavioural Ecology
Center for Non-Linear Phenomena and Complex Systems
CP 231, Campus Plaine
Universite Libre de Bruxelles
Boulevard du Triomphe
1050 Bruxelles
Belgium

Tel: 32-2-650.5776
Fax: 32-2-650.5767
E-mail: sgoss@ulb.ac.be

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

From: Andrew Chipperfield <chip@acse.sheffield.ac.uk>
Date: Mon, 26 Apr 93 10:08:49 BST
Subject: IEE Colloquium: GAs in Control Systems Engineering

Nature has been a source of inspiration and metaphors for powerful new
computing approaches which find ready application in control systems
engineering.  An IEE Colloquium on Genetic Algorithms in Control
Systems Engineering to be held at Savoy Place, London, England on
Friday, 28th May will focus on new work involving one of these
metaphors and its application in control.

The meeting will begin by introducing the topic and reviewing
potential application areas.  Presenters will relate their
experiences in the implementation of genetic algorithms and will
describe a multiobjective optimisation approach to design and the
exploitation of parallel and distributed processing in off-line and
on-line implementations.  Specific applications will include robot
motion planning, identification of partially known systems,
classifier systems, telecommunication routing and a medical
application of multivariable fuzzy control.  Besides contributions
from leading workers in the UK, the programme includes presenters
from France, Germany and Mexico.--

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

From: wsi@vexpert.dbai.tuwien.ac.at (Wolfgang Slany)
Date: Thu, 22 Apr 93 09:48:32 +0200
Subject: FLAI93-exhibition

An exhibition on books will be organized at the forthcoming conference
on "Fuzzy Logic in Artificial Intelligence" (FLAI 93), Linz, Austria.
If you know of any recent books that are of particular interest to the
participants, your suggestions would be of great help. Topics include
Neural Networks, Genetic Algorithms, Expert Systems, Fuzzy Logic
(Theory and Application).

Any suggestions are welcome. Send references to "benedikt@zdvaxa.una.ac.at".
Please, DO NOT use "fuzzy-mail", "NAFIPS-L", "comp.ai.fuzzy", etc.
Thank you in advance              

Best regards, and see you in Linz at FLAI'93,

Josef Benedikt                         Email: benedikt@zdvaxa.una.ac.at

FUZZY LOGIC IN ARTIFICIAL INTELLIGENCE, LINZ (AUSTRIA) June 28-30, 1993
Latest Automatic News:            flai93_info@vexpert.dbai.tuwien.ac.at
Administrative Stuff:             flai93_secr@vexpert.dbai.tuwien.ac.at
Conference Chairpersons:               flai93@vexpert.dbai.tuwien.ac.at
or fax to Ms. Heidi MILOS at                              +43-1-5055304

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


