[comp.ai.neural-nets] Neocognitron

bye@tasman.cc.utas.edu.au (Stephen James Bye) (03/08/91)

I am currently developing a program that simulates the operation of the neocognitron. I have almost completed the task, but have encountered a problem that I have not been able to solve. The network seems to be functioning correctly, but the results obtained are not in accord with those found by Fukushima. I was wondering if anyone out there has managed to get the neocognitron working properly! If so, I would appreciate any correspondence and assistance. Please email any useful response.

===========================================
Stephen Bye,     bye@tasman.eee.utas.edu.au
Electrical & Electronic Engineering
University of Tasmania 
===========================================

bakre@remus.rutgers.edu (Ajay Bakre) (06/02/91)

I am planning to use the Neocognitron (Fukushima) for a project. Some days
back there was a posting inquiring for its implementation. I need the
implementation if it is available on the net or any ftp site. I guess
even its simplified version would suffice ( bearing in mind the
complexity of the model).

I am posting this for a friend, and I would greatly appreciate if
the replies are sent to him. Thanks in advance.

His address is:
e-mail: naik@caip.rutgers.edu