IJCAI—2007  WORKSHOP  PROPOSAL

 

Theme:   COMPLEX  VALUED  NEURAL  NETWORKS  AND  NEURO-COMPUTING:

NOVEL  METHODS, APPLICATIONS  AND  IMPLEMENTATIONS

 

CALL  FOR  PAPERS

 

Researchers  are  invited  to  submit  papers  related  to  the  theme  of  the Workshop. Suggested areas of interest are provided below.

 

Organizers  of  the  Workshop

 

Garimella  Rama  Murthy, International  Institute of  Information  Technology, Hyderabad-500032, India,  E-mail :  rammurthy@iiit.ac.in  Phone: +91-40-23001967 extn:321  (Primary  Contact  Person)

 

Vadrevu  Sree  Hari  Rao,  Jawaharlal  Nehru  Technological  University (JNTU ), Hyderabad-500072, INDIA,   E-mail:  vshrao@yahoo.com  Phone: 9346675550

 

 

1. Technical  Description  of   Workshop:

 A  complex  valued  neural  network  is  a   neural  network  (of  arbitrary  topology)  which  consists  of  complex  valued  input  and/or  weights  and/or  thresholds  and/or  activation  functions.  The  need  for  such  neural  networks  is  very  real. For  instance,  in  electrical  engineering, signals  are  complex  valued.  The  processing  of  such  signals  requires  the  design  and  implementation  of  new  complex  valued  neural  network  architectures.  This  subject  has  been  gaining  increasing  interest  and  significance  in  recent  years.  Indeed  several  interesting  applications  of  the  complex  valued  neural  network  architectures  have  been  discussed in the  following  areas:

·         Optoelectronics,

·         Imaging,

·         Optical  computing

·         Remote sensing,

·         Quantum Neural devices and systems,

·         Intelligent  transport systems,

·         Spatiotemporal  analysis  of  Physiological  Neural  Systems

·         Artificial  Neural  Information  Processing

·         Communication  system  design  (Mobile  channel  equalizer  design)

·         Direction  of  Arrival  Estimation (Signal  Processing)

·         Traffic  Control

·         Robotics

·         Neuron  Dynamics

·         Chaos  in  the  complex  domain

 

The complex-valued neural networks (CVNNs) have been widening the scope of applications not only in electronics and informatics, but also in social systems. One of the most important characteristics of the CVNNs is the proper treatment of complex-amplitude information, e.g., the treatment of wave-related / rotation-related phenomena such as electromagnetism, light waves, quantum waves, oscillatory phenomena even including traffic signal control, and color images processing based on adaptive signal rotation in the RGB space.

2. Reason  For  Interest  in  the  Topic:

It  is  becoming  clear  that  CVNNs  will  be  of  Increasing  importance  in  Future.  The  interest   in  CVNNs  is  becoming  clear  in  the

Number  of   Conferences  devoted  to  them.  For  instance,  recently there were several CVNN Special Sessions (SSs) in, for example,
KES 2001-2003, ICONIP 2002 Singapore, ICANN/ICONIP 2003 Istanbul, and ICONIP 2004 Calcutta, where we had large audience and enjoyed heated and encouraging discussions successfully. The series indicates the increasing importance of this new area.

Now we hold a workshop at  IJCAI-2007 for a wider audience, including  Fuzzy and evolutionary researchers, to develop further investigations both in theoretical and application fields. We invite latest results and discuss prospective possibilities of the CVNNs.
Papers that are, or might be, related to the CVNNs are widely solicited.

Topics include, but are not limited to:

 

·        Complex  Valued  Neuron  Dynamics

·         Complex  Valued  Associative  Memories

·         Chaos  in  the  Complex  Domain

·         Spatiotemporal  CVNN  Processing

·         CVNN- based Two-Dimensional  Information Transform

·         CVNN-based  Periodic  Information  Processing

·         Fourier  Domain  CVNN  processing

·         Phase-sensitive  Radar  Signal  Processing

·         Coherent  Optoelectronic  Applications

·         Speech  and  Ultrasonic  Applications

·         Adaptive  Quantum  Devices

·         Quantum  Neural  Networks

·         CVNNs  in  Colour  Image  Processing

·         CVNNs  in  Traffic  Control

·         CVNNs  in  Robotics

·         Quaternion  and  Clifford  Networks

 

  1. WORKSHOP  AGENDA:  

    2.  

    • Key Note  Speeches  by Dr.SevenBuchloz University of Kiel Germany

     

    • Biography: Sven Buchholz received a diploma degree in computer science and a Ph.D. in computer science (summa cum laude), both
      from Christian-Albrechts-University Kiel, Germany. His PhD thesis "A Theory of Neural Computation with Clifford algebras" has been awarded the Best Ph.D. Thesis of the academic year 2004/2005 by the Technical Faculty of Kiel University. He has authored several publications on the topic of Clifford neural networks. At the moment he is a PostDoc at Cognitive Systems Group in Kiel. His research interest include mathematics of neural networks, machine learning, and multidimensional signal theory. Additionally, to his academic experience he has several years of experience experience as statistical consultant.

    Invited Speakers

    Dr.Prem Kalra ,I I T ,Kanpur

    Dr.Rajat De ,I S I ,Calcutta

 

 The  workshop  will  be  organized  in  a  morning  and  afternoon  session  at  IJCAI-2007.  We  expect  to  select  20  papers  from among  the  submissions. The  workshop  can  be  organized  on  any  one  of  the  days  during  6th  to  12th  January  2007.

 

IMPORTANT  DATES:

                         Last  date  for  submission  of  papers:  15th  September  2006

                         Acceptance  of  Papers                       : 20th October  2006

                         Camera  Ready  Papers                      : 15th November 2006

 

Qualifications:

Prof  Vadrevu Sree Hari  Rao  is  a  well  known  researcher  in  Neural  Networks.  His  research  work  on   Bi-directional  memory  is  well  referenced.  He  is  also  an  editor  of   International  Journal  of  Neural  Systems, World  Scientific

 

Dr. Rama  Murthy, Garimella  discovered  and  formalized  the Theory  of   Multi-Dimensional  Neural  Networks.  It  resulted  in  a  unified  theory  of   control,  communication  and  computation.  He  discovered  and Formalized   a   COMPLEX  VALUED ASSOCIATIVE  MEMORY  on  the Complex-hypercube.  Recently  he  proposed  a  novel  model  of  neuron  and the  complex  valued  neural   networks  based  on  such  a  model.