home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.neural-nets
- Path: sparky!uunet!ukma!usenet.ins.cwru.edu!agate!ames!riacs!danforth
- From: danforth@riacs.edu (Douglas G. Danforth)
- Subject: Re: Kanerva's SDM (was Re: Hopfield Nets / Pros & Cons)
- Message-ID: <1992Nov7.012355.1080@riacs.edu>
- Sender: news@riacs.edu
- Organization: RIACS, NASA Ames Research Center
- References: <RINKUS.92Nov3164452@radar.bu.edu>
- Date: Sat, 7 Nov 92 01:23:55 GMT
- Lines: 48
-
- In <1992Nov4.205034.16263@fripp.ri.cadre.com> umn@cadre.com
- (Uttam M. Narsu) writes:
-
- >rinkus@radar.bu.edu (Gerard J. Rinkus) writes:
- >:
- >: This is from an article entitled, "Comparison between Kanerva's SDM
- >: and Hopfield-type Neural Networks" from Cognitive Science 12, 299-329
- >: (1988) written by James D. Keeler
- >:
-
- >[text elided]
-
- >:
- >: --
- >: Gerard J. Rinkus
- >: Dept. of Cognitive and Neural Systems
- >: Boston Univ.
- >: rinkus@cns.bu.edu
-
- >I haven't heard much about Kanerva's work since I read 'Sparse Distributed Memory' many
- >years ago. What's the consensus on his work? Has anyone implemented an SDM in software or
- >hardware?
-
- >--
- >__________________________________________________________________
- >| Uttam M. Narsu, | EMAIL: umn@cadre.com |
- >| c/o Cadre Technologies Inc, | TEL: (401) 351-5950 |
- >| 222 Richmond St., Providence RI 02906 | FAX: (401) 351-7380 |
- >|________________________________________|________________________|
-
- NASA funded a 5 year project at Ames Research Center, Moffett Field,
- California to investigate applications of Sparse Distributed Memory
- to Autonomous Robots, Speech and character recognition, and weather
- prediction (among others). The results of this work are published
- in a number of technical reports obtainable from RIACS
- (Research Institute for Advanced Computer Science) at Ames. A
- patent has been issued for a variation of the SDM design to Louis
- Jaeckel of the SDM group. Richard Prager in England implemented a
- form of SDM for speech recognition. There have been a number of
- other affiliated groups. A board level hardware implementation
- of the address decoder was constructed at Stanford University under
- the direction of Mike Flynn in the Electrical Engineering department.
-
- The NASA restructuring has curtailed the SDM funding and we are all
- out looking for jobs ...
-
- -- Douglas G. Danforth
-
-