[comp.ai.neural-nets] Adaptive Logic Networks

dwelly@saddle-lk.cs.UAlberta.CA (Andrew Dwelly) (04/24/91)

comp.ai comp.ai.neural-net comp.database can.ai

Readers of this group may be interested to hear that we shall shortly be
releasing some unusual neural net software on the net. In the meanwhile it
is availble by anonymous FTP from menaik.cs.ualberta.ca [129.128.4.241].

Applications we have tried up to now include:

Land Use Classification
Census Analysis
Prosthesis control

...and a number of pattern recognition tasks such as OCR. We feel that the
algorithms are suitable for large scale data analysis (hence the crossposting 
to comp.database), and are now searching for other industrial and business
applications.

The algorithm is a radical departure from normal neural-net techniques
because it is based on the synthesis of an Adaptive Logic Network (ALN).
In the process of learning, the system constructs a boolean digital circuit
to perform the task. Training is rapid, and the execution of the resulting
network is extremely fast (and could easily be turned into hardware for
even more speed).

The software includes a library of routines for the "C" programmer, a small
language for data analysis, documentation (in TeX) and some examples. 

Why are we releasing it ? We are hoping for a wider group of users, comments
on the existing software, and partners for future developments. Please send
all mail on the subject to:

arms@cs.ualberta.ca or dwelly@cs.ualberta.ca

Bill Armstrong.
Andrew Dwelly.
Rolf Manderscheid.
--
******************************************************************************
Andy Dwelly : dwelly@cs.ualberta.ca, Tel: 403-492-7591
!@#$%$#, %^&*%, - Listen, who swears ? Christopher Robin has fallen downstairs.
******************************************************************************