Evolving Connectionist Systems
This site is dedicated to Evolving connectionist Systems (ECoS). ECoS
are a family of
constructive artificial neural network algorithms that were first
proposed by Professor Nik Kasabov in 1998 and have since been adopted by many other
researchers. This site is my attempt at bringing together links to
ECoS-related publications, software (including the ECoS Toolbox) and resources. For an overview of ECoS
technology, please see the following paper:
Watts, M.J. A decade of Kasabov’s Evolving Connectionist Systems: A
Review. IEEE Transactions on Systems, Man and Cybernetics Part C -
Applications and Reviews (2009) 39(3) 253-269 (PDF reprint).
There are three major sections to this site: Algorithms, which describes the major ECoS algorithms; Literature, which provides a comprehensive listing of publications relevant to ECoS; and Software, which links to the ECoS software that is available and is home to the ECoS Toolbox.
5 May 2017 - Updated the Applications literature list.
8 March 2016 - Version 1.24 of the ECoS Toolbox released.
5 February 2016 - Version 1.22 of the ECoS Toolbox released.
8 October 2015 - Version 1.2 of the ECoS Toolbox released.
8 April 2012 - Restructured the entire site. Equations are now rendered using MathML.
4 April 2012 - Submission deadline is now 14 May 2012 for the special issue of Evolving Systems on Applications of Evolving Connectionist Systems (CFP)
20 March 2012 - Updated the ECoS Toolbox
24 February 2012 - Small bug fixes in the TrainEFuNN and TrainSECoS applications in the ECoS Toolbox
15 November 2011 - Call for papers: Special Issue of Evolving Systems on Applications of Evolving Connectionist Systems
Maintained by Michael J. Watts