Connectionists: New paper on using deep sparse distributed representations for event learning and recognition

[Rod Rinkus]

Dear Connectionists,

I have a new paper
<http://journal.frontiersin.org/Journal/10.3389/fncom.2014.00160/abstract>
"Sparsey^TM: event recognition via deep hierarchical sparse distributed
codes" in Frontiers in Computational Neuroscience, which I think will be of
general interest to this group.  Many existing hierarchical models propose
localist representations in each sub-region of each hierarchical level.  In
contrast, in Sparsey, all sub-regions (macrocolumns) in all of internal
levels use sparse distributed codes, which confers great computational
efficiency, as we argue in the paper.  This animation
<http://www.sparsey.com/index.html> of a memory trace unfolding in a
6-level Sparsey model shows the proposed general correspondence to cortex
(the levels of macs would correspond to V1, V2, etc.).  I look forward to
any questions/feedback from the community.

Sincerely,
Rod Rinkus


-- 
Gerard (Rod) Rinkus, PhD
President,
rod at neurithmicsystems dot com
Neurithmic Systems LLC <http://sparsey.com>
275 Grove Street, Suite 2-400
Newton, MA 02466
617-997-6272

Visiting Scientist, Lisman Lab
Volen Center for Complex Systems
Brandeis University, Waltham, MA
grinkus at brandeis dot edu
http://people.brandeis.edu/~grinkus/
<http://people.brandeis.edu/%7Egrinkus/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20141219/03dc8567/attachment.html>