🚀 go-pugleaf

Neuron Digest   Sunday,  6 Apr 1997                Volume 97 : Issue 17

Today's Topics:
  NEW MIT BOOK! Mathematical Methods for Neural Net Analysis and Design
                      Summer School info (corrected)
           European Workshop on Neuromorphic Systems: final CFP
              New Home for The Genetic Programming Notebook
     ANNOUNCING -- A New E-List on the Writings of Friedrich A. Hayek
                           Course announcement
           Research jobs in neural nets and pattern recognition
              Thesis on data exploration with SOMs available


Send submissions to "neuron@psych.upenn.edu". Send administrative
questions and [un]subscribe requests to "neuron-request@psych.upenn.edu".
The ftp archives (including back issues) are available from
psych.upenn.edu in pub/Neuron-Digest or by sending a message to
"archive-server@psych.upenn.edu".

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

Subject: NEW MIT BOOK! Mathematical Methods for Neural Net Analysis and Design
From:    Richard M Golden <golden@utdallas.edu>
Date:    Tue, 11 Mar 1997 12:48:32 -0600

     Mathematical Methods for Neural Network Analysis and Design
                        Richard M. Golden

MIT Press (1996)                ORDERING INFO:
ISBN 0-262-07174-6              http://www.amazon.com              
Cloth ($65.00), 432 pages.      http://www-mitpress.mit.edu/order-info.html
                                1-800-356-0343 (MIT BOOK ORDER Department)
***FOR MORE INFO: http://www.utdallas.edu/~golden/book_abs.html   

This textbook teaches students how to carefully use a powerful set of 
mathematical tools for analyzing and designing a wide variety of 
NONLINEAR HIGH-DIMENSIONAL Artificial Neural Network (ANN) systems.

Chapter 1: ANN systems with Neuroscience, Psychology, Engineering Applications
Chapter 2: Specific ANN system architectures; classification/learning paradigms 
Chapter 3: LaSalle's Invariant Set Theorem for behavioral analysis of ANNs
Chapter 4: Stochastic Approximation Theorem for behavioral analysis of ANNs
Chapter 5: Nonlinear Optimization Theory for ANN system design 
Chapters 6,7: Bayesian Decision Theory and Markov Random Fields for the
         design of "rational" ANN classification/learning objective functions.  
Chapter 8: Confidence intervals for an ANN system's predictions. Statistical 
           tests for: (i) pruning/adding units, and (ii) model selection.  
Solutions: Solutions to over 100 ANN system analysis and design problems   

***** This message is being sent to multiple mailing lists. My apologies if
      you receive this message more than once. ******  

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

Subject: Summer School info (corrected)
From:    Marco Gori <marco@McCulloch.ing.UNIFI.IT>
Organization: DSI - AI Lab - University of Florence (Italy)
Date:    Mon, 24 Mar 1997 18:13:34 +0100


                  ================================         
                    CAIANIELLO SUMMER SCHOOL 
                                 on
                  Adaptive Processing of Sequences
                  ================================

                  Vietri sul Mare, Salerno (Italy)
                         6-14 September 1997

                             Directors

                      C.L. Giles and M. Gori
    

                             Lectures

- --------------------------------------------------------------------------
C.L. Giles (NEC, USA)             M. Gori (Siena Univ. IT)
A.C. Tsoi (Wollongong Univ. Aus)  P. Baldi (Caltech, USA)
Y. Bengio (Montreal Univ. CA)     P. Frasconi (Florence Univ. IT)
H. Bourlard (IDIAP, CH)           H. Siegelmann (Technion Univ. Israel)
P. Gallinari (Curie Univ. FR)     Z. Ghahramani (Toronto Univ. CA)      
B.A. Pearlmutter (Salk Ins., USA) E. Wan (Oregon Inst. USA)
M. Mozer (Colorado Univ. USA)     A. Sperduti (Pisa Univ. IT)
- ---------------------------------------------------------------------------

                            Objectives

This summer school is intended to provide a unified view of different
techniques for processing temporal information that rely hardly on
learning.  The topics of the school include architectures (e.g., recurrent
networks, Markovian models), optimization algorithms (e.g., gradient-based,
EM-based), theoretical results, integration with symbolic systems, and
applications (control, speech recognition, time series analysis) related to
sequence processing tasks.

                             Sponsors

AI*IA   (Italian Association for Artificial Intelligence)       
CNR     (Italian Scientific Research Council)
IEEE Neural Networks Council (Italian RIG)
IIASS   (International Institute for Advanced Scientific Studies)
SIREN   (Italian Neural Networks Society)
University of Salerno
University of Siena

                       General Information

This is the second Summer School dedicated to the memory of Prof. E.R.
Caianiello. Prof. M. Marinaro (Univ. of Salerno and IIASS) and Prof. M.
Jordan (MIT, Dept. of Brain and Cognitive Sciences) are the organizers and
advisors of this cycle of schools. The first school was organized last year
in Erice (Italy) on graphical models by M. Jordan and D. Heckerman.
  
People wishing to attend the school should send an e-mail to
Prof. Marco Gori (marco@mcculloch.ing.unifi.it) and then
send the written application to:

Prof. Maria Marinaro - IIASS "Eduardo R. Caianiello"
via Pellegrino, 19 - 84019 Vietri sul Mare (SA) Italy,
Tel: +39 89 761167 - Fax: +39 89 761189

They should specify:
1) date and place of birth together with present nationality;
2) degree and other academic qualifications;
3) present position and place work.

Young researchers with little experience should include a letter of
recommendation from the head of their research group or from a senior
scientist, active in the field.  The total fee, which includes full board
and lodging (arranged by the school), is $1000 US. Thanks to the generosity
of the sponsoring Institutions, partial support can be granted to some
deserving students who need financial help. Requests to this effect must be
specified and justified in the application letter. Closing date for
application: June 15th, 1997

More information can be found at http://www.ing.unisi.it/DII/research/neural/SUMMER_SCHOOL

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

Subject: European Workshop on Neuromorphic Systems: final CFP
From:    Leslie S Smith <Leslie.Smith@ee.ed.ac.uk>
Date:    Tue, 25 Mar 1997 13:54:41 +0000

EWNS: 1st European Workshop on Neuromorphic Systems
- ---------------------------------------------------
Final Call for Papers
- ---------------------

29-31 August 1997, University of Stirling, Stirling, Scotland

Organisers: 

Centre for Cognitive and Computational Neuroscience
University of Stirling

and 

Department of Electrical Engineering
University of Edinburgh

Neuromorphic systems are implementations in silicon of sensory and neural 
systems whose architecture and design are based on neurobiology. This growing 
area proffers exciting possibilities such as  sensory systems which can  
compete with human senses and pattern recognition systems that can run in 
real-time. The area is at the intersection of many disciplines: 
neurophysiology, computer science and electrical engineering.

Papers are requested in the following areas:

Design issues in sensorineural neuromorphic systems: 
auditory, visual, olfactory, proprioception, sensorimotor systems.

Designs for silicon implementations of neural systems.

Papers not exceeding 8 A4 pages are requested: please send 3 copies.
These should be sent to 

Dr. Leslie Smith
Department of Computing Science
University of Stirling
Stirling FK9 4LA
Scotland

email: lss@cs.stir.ac.uk
FAX (44) 1786 464551

We also propose to hold a number of discussion sessions on some of the 
questions above. Short position papers (less than 4 pages) are requested.

Key Dates
- ---------
Submission Deadline: Mon 7th April 1997
Notification of Acceptance: June 2nd 1997

World Scientific will be publishing a book based on this conference.

Applications are invited from UK Ph.D. students to cover the cost of travel 
and registration. Please contact Dr. Leslie Smith.

Full information is on the WWW at

http://www.cs.stir.ac.uk/~lss/Neuromorphic/Info1.html

___________________________________________________________________________

Registration form:

Full name

Full postal address



Telephone number

Fax number

email address



Registration fee:                  UKpounds 100             _____
(includes lunches and coffees 
and copy of proceedings)
Conference Dinner                  UKpounds 25             _____

Accommodation required:

28 August                          UKpounds 17.02          _____
29 August                          UKpounds 17.02          _____
30 August                          UKpounds 17.02          _____
31 August                          UKpounds 17.02          _____
1 September                        UKpounds 17.02          _____



Total                                                      _____

Cheques should be made payable to the University of Stirling. Eurocheques 
are acceptable. Unfortunately the University does not take credit cards.

Please send form with remittance to

Dr. Leslie Smith,
Department of Computing Science and Mathematics,
University of Stirling,
Stirling FK9 4LA
Scotland, UK.

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

Subject: New Home for The Genetic Programming Notebook
From:    Jaime Fernandez <jjf@mickey.jsc.nasa.gov>
Organization: Metrica
Date:    Tue, 25 Mar 1997 17:06:51 -0600

The Genetic Programming Notebook has a new home.  It is at:
http://www.geneticprogramming.com/

The GP Notebook site contains information on the following categories:

      Genetic Programming

 GP Tutorial
 Software
 People
 Other Sites
 Research Groups
 Misc
 Bibliographies
 Papers
 Journals
 FAQ
 Calls for Papers
 Conferences
 Commercial


         Genetic Algorithms

 Genetic Algorithms
 Software
 People
 Other Sites
 Research Groups 
 Misc 
 Bibliographies 
 Papers 
 Journals 
 FAQ 
 Courses 
 Parallel 
 Repositories 
 Tutorials 


     Artificial Intelligence & Robotics

 Artificial Intelligence 
 FAQs
 Newsgroups
 Machine Learning 
 Artificial life Sites 
 Fuzzy Logic 
 Neural Nets 
 Robots 
 Financial 
 Programming 




Jaime Fernandez


- -- 
Jaime J. Fernandez Jr.   |  Email:   jjf@jjf.com
Metrica, Inc.            |  WWW:     http://www.jjf.com
Author of the GP Notebook:   http://www.geneticprogramming.com/
And Castles of the World:    http://www.castles.org/


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

Subject: ANNOUNCING -- A New E-List on the Writings of Friedrich A. Hayek
From:    GBRansom@aol.com
Date:    Wed, 26 Mar 1997 02:07:32 -0500

Hayek-L on listserv@maelstrom.stjohns.edu

   HAYEK-L@MAELSTROM.STJOHNS.EDU is an international network for the
   discussion of the ideas of Friedrich A. Hayek. Hayek-L is
   intended as a resource for scholars and others doing research
   connected to the contributions of Friedrich Hayek.  Hayek 
   is the co-originator of the Hebb-Hayek synaptic learning model,
   and author of the well known _The Sensory Order_.  Hayek's
   work in neuroscience has gained prominence in recent years due 
   to Gerald Edelman's _Neural Darwinism_, and Joaquin Fuster's 
   _Memory in the Cerebral Cortex:  An Empirical Approach to Neural
   Networks in the Human and Nonhuman Primates_. 

   The basic purpose of the list is to serve as a forum for scholarly
   discussions and as a clearing house the distribution of
   information on academic conferences, publication opportunities,
   fellowship information, academic grants, and job openings of
   interest to Hayek scholars.  Subscribers are encouraged to post
   questions, comments, or announcements of interest to individuals
   working on topics related to Hayek's writings. Appropriate
   postings might pertain to work currently in progress, the
   development of course materials, bibliographical material of
   interest to Hayek scholars, useful internet resources, etc.

   The list is for scholars and others interested the ideas of
   Friedrich A. Hayek without restriction according to interest or
   professional affiliation. Hayek-L is not devoted to any particular
   niche within the scope of Hayek's oeurvre, but instead welcomes
   contributions on any aspect of the full range of Hayek's
   contribution to contemporary scholarship.  Discussions of the
   scholarly contributions of other important thinkers who have
   developed or criticized aspects of Hayek's work are also welcomed.
   E.g. discussions of the work of Gerald Edelman, Karl Popper,
   Israel Kirzner, Ronald Dworkin, Walter Weimer, Robert Nozick and
   others related to the work of Friedrich Hayek are also welcome.

   To subscribe to Hayek-L, send mail to
   listserv@maelstrom.stjohns.edu  with the message (body):
 
   SUBSCRIBE HAYEK-L your full name

   For example:  SUBSCRIBE HAYEK-L Max Doe

   Owner:  Greg Ransom  <Gregransom@aol.com>
           Juan Carlos Garelli  <lagare@attach.edu.ar>
           HAYEK-L list coordinators


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

Subject: Course announcement
From:    Tom Vogl <tvogl@wo.erim.org>
Date:    Tue, 01 Apr 1997 14:57:54 -0500


                            George Mason University
                                   ANNOUNCES

                      COMPUTER MODELING OF BIOLOGIGAL SYSTEMS

          A ONE WEEK intensive course devoted to the design, analysis,
       evaluation, and interpretation of computer models with emphasis on
                         the modeling of biological systems. 

          Monday, June 2 through Friday June 6, 1997, 8:30a.m. - 5:00p.m. 
                                                               
                George Mason University,  Prince William Campus

     7946 Donegan Drive,  Manassas, Virginia 20109-2870 (703) 330-5620



                          FOR WHOM THE COURSE IS INTENDED

Scientists and engineers who are interested or involved in the writing of
computer models of natural phenomena; Managers responsible for the
specification or development of such computer models; All those
individuals who are, or are considering, applying, specifying, or funding
computer models of natural phenomena.

                                ABOUT THE COURSE

The course is designed to address the needs of those whose task it is to
critically analyze, evaluate, and interpret models developed by others or
to design and develop models by themselves.  The emphasis of the course is
on model design, analysis, and evaluation; on recognition of the strengths
and limitations of various modeling approaches, and on the interaction of
client expectations on model design.  Effectiveness of written and oral
presentation of models and modeling results will be stressed.  Development
of computer code is optional.  One week before class begins, a short
written assignment will be required of all participants.  Although no
formal homework will be assigned during the course, students should be
prepared to do independent reading and writing, since an important
component of the course will be the description, analysis, and design of a
model chosen for development by each student based on their individual
interests.  In order to permit extensive interaction among class
participants, class size will be limited to 20 students. Guest lecturers
will describe models that they have designed and implemented and the
pitfalls that they encountered.

Prerequisites: A bachelor degree and familiarity with the concepts of
biological systems, calculus/differential equations, and computer
programming, or consent of the instructor.  Extensive skills in all of
these areas are not presupposed.  Students unsure about the demands and
aims of the course are urged to confer with the instructor, preferably by
e-mail (tvogl@gmu.edu) before registering.

Text: Mathematical Models on Biology by L. Edelstein-Kechet, McGraw-Hill,
1988).

                            INSTRUCTORS

Dr. Thomas P. Vogl, MBTI/GMU.  Dr. Vogl has been active in the development
of models and computer aided design for over 30 years. He is the author of
over 130 papers and book chapters.  His current research interests focus
on the modeling of information processing in the brain.  For further
information see http://mbti.gmu.edu/FACULTY.html. He will be assisted by
Dr. Alexander Golovan, Senior Researcher and Group Leader, Laboratory for
Neuroinformatics of Sensory and Motor Systems, A. B. Kogan Research
Institute for Neurocybernetics, Rostov State University,
Rostov-on-the-Don, Russia and Post-Doctoral Fellow at MBTI/GMU.

REGISTRATION INFORMATION

George Mason University students who desire to take the course for credit
(3 credits) should register for CSI 739-401. Regular tuition rates
apply. This can be done through regular registration channels after
obtaining permission from the instructor. (703)528-5250 Fax(703)524/3527,
email:tvogl@gmu.edu

For others, the cost of the course is $950.00.  The cost includes course
materials and continental breakfasts.  To register, complete the attached
registration form and return with your check payable to George Mason
University, purchase order, or Visa/Master Card authorization to: 

George Mason University
Prince William Institute
7946 Donegan Drive
Manassas, VA 20109-2870
ATTN: Computer Modeling of Biological Systems

By Fax: If you have a Visa or MasterCard, you may register by FAX. 
Just fill out the registration form, including your credit card number, and
send it via FAX to (703) 330-0683.  Follow-up by mailing the enrollment
form to confirm Visa/MasterCard registration.  Registrations will be
confirmed by mail.

Please be sure to include your e-mail address so that information can be
sent to you in advance of the course.

For further information regarding registration, please call (703) 330-5620

____________________________________________________________________________

                                 REGISTRATION FORM

Name___________________________________

Title__________________________________

Organization___________________________

Address________________________________

City___________________________________

State___________________Zip____________

Day Phone______________________________

Evening Phone__________________________

FAX____________________________________

e-mail_________________________________

Soc. Sec.#_____________________________


Fee: $950.00

Amount Paid____________________________

Check Enclosed Payable to George Mason

University (Federal I.D. #54-0836354)

 By Visa, MasterCard, or  PO:

Card or PO number  ____________________

Amount Charged_________________________

Credit CardNumber______________________

Expiration Date________________________

Cardholder Name________________________

Date___________________________________

Authorizing Signature__________________
 
  




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

Subject: Research jobs in neural nets and pattern recognition
From:    w.penny@ic.ac.uk
Date:    Thu, 03 Apr 1997 15:22:20 +0100

THREE POST-DOCTORAL RESEARCH POSITIONS
IN PATTERN RECOGNITION  / NEURAL NETWORKS RESEARCH

Three post-doctoral research positions are available within the Neural
Systems Section of the Department of Electrical & Electronic Engineering to
work on the theory and application of advanced pattern recognition
techniques, in particular the use of Bayesian methods and neural networks.
Two positions are funded for two years and the third nominally for three
years with a yearly evaluation.  All projects involve research in
statistical pattern recognition with applications in the biomedical field.
Experience in pattern recognition and Bayesian statistics would be an
advantage.  A good understanding of data processing (especially signal
processing) techniques is desired as is experience of UNIX, C and Matlab.
The positions are funded by the Jefferiss Research Trust, the European
Commission and British Aerospace plc respectively. The salary scale will be
RA1A , GBP 14,732 - 22,143 per annum (exclusive of London Allowance of GBP
2,134) depending on age and experience.  Further information may be
obtained from http://www.ee.ic.ac.uk/research/neural/positions.html or via
e-mail to Dr Stephen Roberts (s.j.roberts@ic.ac.uk).  The closing date for
applications is April 11th 1997.

In recent years, great interest has developed in the use of non-classical
methods for statistical analysis of data as part of a general increase
towards the use of artificial intelligence methods.  One genre which has
shown itself to be particularly suitable is that of connectionist models, a
subset of which are referred to as a artificial neural networks (ANNs).
Classical statistical methods rely upon the use of simple models, such as
linear or logistic regression, in order to 'learn' relationships between
variables and outcomes. ANNs offer a far more flexible model set, indeed it
has been shown that they have the property of universal approximation so
they are able, in principle, to estimate any set of arbitrary relationships
between variables.  Furthermore, they may model non-linear coupling between
sets of variables.  Part of the momentum of the recent development of ANNs
for pattern recognition, regression and estimation problems must be
attributed to the manner in which ANNs conform to many of the traditional
statistical approaches, i.e. they may estimate Bayesian probablilities in
the case of classification and conditional averages in the case of
regression.

1) The use of Neural Networks to Predict the Development and Progression of
Kaposi's Sarcoma (KS).

This is a joint project funded by the Jefferiss Research Trust between the
Department of Electrical and Electronic Engineering, Imperial College of
Science, Technology & Medicine and the Department of Genito-urinary
Medicine, St. Mary's Hospital.

Kaposi's sarcoma (KS) is a vascular tumour, which is more common and often
aggressive in patients with underlying immunosuppression (post-transplant
KS and AIDS-associated KS).  KS was first described by the Hungarian
pathologist Moritz Kaposi in 1872, yet still remains something of a
clinical enigma, being an unusual tumour of unknown origin.

The aim of this research is to determine factors that influence the
variable progression rate of KS in HIV infected individuals. There is
currently no means of predicting which patients will develop KS and no
understanding of the relationship between the forms of the disease. The aim
of the project is to carry out multi-variable analyses in order to define
clinical end-points and provide guidelines for better patient management.

A number of variables will be available to the system.  The reliability and
utility of each with regard to the prediction of patient outcome, however,
is generally unknown.  Classical regression analysis offers some powerful
methods of selection and ranking within a subset of features or variables.
Whilst such methods should be used for completeness and comparison, it is
noted that recent developments in Bayesian learning theory have offered the
possibility to assess the utility of variables from within the ANN
structure.  Each input variable has a separate weighting factor, or in
Bayesian terminology, a hyper-prior, associated with it.  This technique
has become known as automatic relevance determination or ARD.  Such an
assessment is devoid of the strong assumptions of independence and
linearity of most of the classical regression methods.

It is feasible for an ANN to produce not only a set of output variables
(predictions or classifications, for example) but also an associated set of
confidence or validation measures (describing the probable error on each
output).  This enables the tracking of predictions of future events in a
more robust framework and furthermore allows for the accurate fusion of
information from more than one source and the incorporation of temporal
information, i.e. poor quality information from the present time may be
suppressed in favour of more reliable information from past or future as it
becomes available.

If we may regard the system as aiming to produce a probability distribution
in some 'outcome space', then several possible approaches to analysis are
made available.  As temporal information is retained (i.e. outcomes are
based upon the entire course of the patient's history, not just present
information) we may seek information regarding the effect of each piece of
information (test result or partial diagnosis) on the probability
distribution in the 'outcome space'.  Two pieces of information may be
obtained from this approach. How important a partial decision or test
result is to the probability of certain outcomes and how important it is to
changing the uncertainty we have in the outcome results.  Clearly, the goal
will be to indicate tests and/or procedures which not only increase the
survivability probabilities but also make the estimated outcomes less
variant, so we have more confidence in the predictions (this means not only
increasing the height of a favourable node in the posterior probability
space, but also attempting to reduce the variance of the distribution).  In
order to accommodate for multiple output hypotheses we propose to utilise a
procedure similar to that detailed in (Bishop 1995) whereby the output
distribution is modelled multi- modally.  This has the added benefit that
individual modes (possible outcomes) may be tracked separately.  This
representation is also similar to that taken in a mixture of experts
approach.

REFERENCES 

1. Bishop CM. Neural Networks for Pattern Recognition. Oxford University
Press, Oxford, 1995.

2. Ripley BD. Pattern Recognition and Neural Networks. Cambridge University
Press, Cambridge, 1996.

3. Roberts SJ and Penny W. Novelty, Confidence and Errors in Connectionist
Systems.  Proceedings of IEE colloquium on fault detection and intelligent
sensors, IEE, September 1996.

4. Penny W and Roberts SJ. Neural Networks with Error Bars. Departmental
report, also submitted to IEEE transactions on neural networks, February
1997, available from http://www.ee.ic.ac.uk/staff/hp/sroberts.html


2)      SIESTA (EU funded project)

Siesta is a an EU funded project which involves Imperial College and 10
other European partners.  The aim of the project is to define and produce a
system which is capable of continuous evaluation of the state of the brain
during the sleep-wake cycle.  Such an automated system is of enormous value
in the clinical field and the research into multi-channel signal
processing, fusion and pattern recognition form a challenge to the most
modern techniques.  The state of the brain will, primarily, be monitored
via its electrical activity (the EEG).

One of the most well-known approaches from literature to achieve a
continuous description of EEG state is the system developed by Roberts &
Tarassenko (1992a, 1992b). This approach will be used as a general basis
for the research in SIESTA.  Roberts & Tarassenko (henceforth, R&T') used
a self-organizing feature map (SOM) to perform unsupervised topographic
mapping of feature vectors consisting of 10 coefficients of a Kalman filter
algorithm applied to the raw EEG. This self- organizing network discovered
eight distinct clusters in which the brain state remained preferentially.
>From the general trajectories between these halting states, R&T discovered
three distinct patterns which roughly corresponded to the global brain
states W (wakefulness), S (slow wave or deep sleep) and R (REM sleep). The
outputs from the self-organizing feature map were subsequently mapped via a
Radial Basis Function (RBF) classifier onto three outputs, trained by
sections of data on which experts agreed upon the stage W, S, or R. For
each input, the resulting network produced probabilities for the three
global stages, describing intermediate stages as a combination of three
'mixing fractions' .

This general approach, yielding a novel way of describing brain state while
exploiting some of experts' knowledge in a partly supervised method, will
be adopted and extended in the following ways

  Many features extracted from the signals will be considered.
 
  Instead of the 2-dimensional feature map used by R&T, alternative
approaches will be investigated.  It has been shown that combinations of
other clustering and mapping methods can outperform SOMs. Moreover, since
topographic mapping is only exploited for visualization, the general
approach can be based on more advanced clustering techniques (e.g.
techniques for non-Gaussian clustering (Roberts 1997) or Bayesian-inspired
methods).
 
  in order to cope with the large number of input features to be
investigated active feature selection methods will be applied.
 
  techniques for intelligent sensor fusion will be investigated. When
multiple sources are combined to lead to classification results, it is not
trivial to decide which are the most relevant sources at any given time, or
what should happen when sources fail to provide input (e.g. because an
electrode is faulty). Approaches based on the computation of running error
measures can be employed here.

The Imperial College group will form be a leading centre in the theory
subgroup of the project.  We will be active in the researching of

  Mixture density networks and mixtures of experts
  Model estimation and pre-processing
  Active sensor fusion
  Active feature and data selection
  Unsupervised data partitioning methods (clustering)
  Model comparison and validation


References

1. S.J. Roberts, Parametric and Non-parametric Unsupervised Cluster
Analysis.  Pattern Recognition, 30 (2) ,1997.
 
2. S.J. Roberts and L. Tarassenko, The Analysis of the Sleep EEG using a
Multi- layer Neural Network with Spatial Organisation. IEE Proceedings Part
F, 139(6), 420-425, 1992a.
 
3. S.J. Roberts and L. Tarassenko, New Method of Automated Sleep
Quantification.  Medical and Biological Engineering and Computing, 30(5),
509-517, 1992b.


3) Assessment of cortical vigilance

This project, funded by British Aerospaces Sowerby Research Centre at
Bristol, aims to assess and predict lapses in vigilance in the human brain.
Recordings of the brains electrical activity are to be recorded and
analysed.

The utility of a device or system which may monitor an individual's level
of vigilance is clear in a range of safety-critical environments.  We
propose that such utility would be enhanced by a system which, as well as
monitoring the present state of vigilance, made a prediction as to the
likely evolution of vigilance in the near future.  To perform both these
tasks, i.e. a static pattern assessment and a dynamic tracking and
prediction, sophisticated methods of information extraction, sensor fusion
and classification/regression must be employed.  Over the last decade the
theory of artificial neural' networks has been pitched within the
framework of advanced statistical decision theory and it is within this
framework which we intend to work.

The aim of the project is to work towards a practical real-time system.
The latter should be minimally intrusive and should make predictions of
future vigilance states.  Where appropriate, therefore, the investigation
will assess each technique in the developing system with a view to its
implementation in a real-time environment.

The project will involve research into :

  New methods of signal complexity and synchronisation estimation
  Information flow estimation in multi-channel environments
  Active sensor fusion
  Prediction and classification
  Error estimation
  State transition detection and state sequence modelling

Reference

Makeig, S. and Jung, T-P. and Sejnowski, T. (1996), Using feedforward
neural networks to monitor alertness from changes in EEG correlation and
coherence, Advances in Neural Information Processing Systems (NIPS), MIT
Press, Cambridge, MA.
 

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

Subject: Thesis on data exploration with SOMs available
From:    Sami Kaski <sami@guillotin.hut.fi>
Date:    Fri, 04 Apr 1997 14:56:49 +0200


The following Dr.Tech. thesis is available at

http://nucleus.hut.fi/~sami/thesis/thesis.html (html-version)
http://nucleus.hut.fi/~sami/thesis.ps.gz       (compressed postscript, 300K)
http://nucleus.hut.fi/~sami/thesis.ps          (postscript, 2M)


The articles that belong to the thesis can be accessed through the page

http://nucleus.hut.fi/~sami/thesis/node3.html

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

Data Exploration Using Self-Organizing Maps

Samuel Kaski

Helsinki University of Technology
Neural Networks Research Centre
P.O.Box 2200 (Rakentajanaukio 2C)
FIN-02015 HUT, Finland


Finding structures in vast multidimensional data sets, be they
measurement data, statistics, or textual documents, is difficult and
time-consuming. Interesting, novel relations between the data items
may be hidden in the data. The self-organizing map (SOM) algorithm of
Kohonen can be used to aid the exploration: the structures in the data
sets can be illustrated on special map displays.

In this work, the methodology of using SOMs for exploratory data
analysis or data mining is reviewed and developed further. The
properties of the maps are compared with the properties of related
methods intended for visualizing high-dimensional multivariate data
sets. In a set of case studies the SOM algorithm is applied to
analyzing electroencephalograms, to illustrating structures of the
standard of living in the world, and to organizing full-text document
collections.

Measures are proposed for evaluating the quality of different types of
maps in representing a given data set, and for measuring the
robustness of the illustrations the maps produce.  The same measures
may also be used for comparing the knowledge that different maps
represent.

Feature extraction must in general be tailored to the application, as
is done in the case studies. There exists, however, an algorithm
called the adaptive-subspace self-organizing map, recently developed
by Kohonen, which may be of help. It extracts invariant features
automatically from a data set. The algorithm is here characterized in
terms of an objective function, and demonstrated to be able to
identify input patterns subject to different transformations.
Moreover, it could also aid in feature exploration: the kernels that
the algorithm creates to achieve invariance can be illustrated on map
displays similar to those that are used for illustrating the data
sets.

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

End of Neuron Digest [Volume 97 Issue 17]
*****************************************