milo@ndmath.UUCP (09/02/87)
Ok, here's a quick question for anyone who's getting into Neural Networks. If you setup the type of network described in BYTE this month, or the type used in the program recently posted to the net, what happens if you feed it an input image that is not aligned right? For example, in the Byte article they demonstrate correct recall of an image corrupted by randomly flipping a number of bytes, simulating "noise". What would happen if they just shifted the input image one or two bits to the left? Would the network still recognize the pattern? Greg Corson ...seismo!iuvax!ndmath!milo
mikek@boulder.Colorado.EDU (Mike Kranzdorf) (09/03/87)
I am not familiar with the net in Byte, but I assume it is a two layer net, like the one that was posted. If this is the case, shifted patterns will not be recognized. It takes at least three layers for a net to have an internal representation of the structure of an input pattern. A good overview paper describing these kinds of conditions can be found in the IEEE ASSP (Acoustics, Speech, and Signal Processing) Magazine April 1987, Volume 4, Number 2, "An Introduction to Computing with Neural Nets" by Richard P. Lippmann. The article focuses on catagorizers, but is informative about nets in general. --mike
smoliar@vaxa.isi.edu (Stephen Smoliar) (09/04/87)
In article <277@ndmath.UUCP> milo@ndmath.UUCP (Greg Corson) writes: >Ok, here's a quick question for anyone who's getting into Neural Networks. >If you setup the type of network described in BYTE this month, or the >type used in the program recently posted to the net, what happens if you >feed it an input image that is not aligned right? > >For example, in the Byte article they demonstrate correct recall of an image >corrupted by randomly flipping a number of bytes, simulating "noise". What >would happen if they just shifted the input image one or two bits to the left? >Would the network still recognize the pattern? > After after that, what happens if you change the SCALE of the image, expanding or contracting it? What sort of data base of images would be required for recognition by an ambulatory robot (or automated land vehicle, if you prefer)?
geb@cadre.dsl.PITTSBURGH.EDU (Gordon E. Banks) (09/09/87)
In article <3523@venera.isi.edu> smoliar@vaxa.isi.edu.UUCP (Stephen Smoliar) writes: >In article <277@ndmath.UUCP> milo@ndmath.UUCP (Greg Corson) writes: >>Ok, here's a quick question for anyone who's getting into Neural Networks. >>If you setup the type of network described in BYTE this month, or the >>type used in the program recently posted to the net, what happens if you >>feed it an input image that is not aligned right? I didn't see the Byte article, but the simple neural networks that I have seen (such as the one that solves the T-C problem by Hinton & Rummelhart in the PDP book) do not generalize very well. You can train the hidden units with a given input, but then if you shift the pattern, it won't work. I asked Rummelhart about this, and he said that once the hidden units develop the patterns (such as edge detectors and center-surround, etc.) you do not need to retrain for each translation of the pattern, but you need to add more units to the network. These units have the same weights as the previously trained units, but they have a different field of view. You have to have another set of units for each region which can possibly contain the image. Alternatively, you have to have a scheme for making sure the image is "centered" in the field of view. Sounds like there is some room for interesting research here, maybe a thesis.
g451252772ea@ucdavis.UUCP (g451252772ea) (09/13/87)
> overview paper describing these kinds of conditions can be found in the > IEEE ASSP (Acoustics, Speech, and Signal Processing) Magazine April 1987, > Volume 4, Number 2, "An Introduction to Computing with Neural Nets" by > Richard P. Lippmann. The article focuses on catagorizers, but is > informative about nets in general. > > --mike I found the 4/87 issue (and the rest of 1987) , but not this article. Are you certain of this reference? Thanks... Ron Goldthwaite, UC Davis Animal behavior and Psychology 'Economics is a branch of ethics, pretending to be a science... ethology is a science, pretending relevence to ethics.'
finegan@uccba.UUCP (09/24/87)
In article <759@ucdavis.UUCP>, g451252772ea@ucdavis.UUCP (g451252772ea) writes: > > IEEE ASSP (Acoustics, Speech, and Signal Processing) April 1987, > > I found the 4/87 issue (and the rest of 1987) , but not this article. > Are you certain of this reference? Thanks... > I am not sure if it was April (I believe it was), but the whole journal is devoted to the subject of Neural Nets for that issue, and definitely exists. - Mike Finegan ...!{hal|pyramid}!uccba!finegan
mikek@boulder.Colorado.EDU (Mike Kranzdorf) (09/24/87)
In article <1241@uccba.UUCP> finegan@uccba.UUCP (Mike Finegan) writes: >In article <759@ucdavis.UUCP>, g451252772ea@ucdavis.UUCP (g451252772ea) writes: >> > IEEE ASSP (Acoustics, Speech, and Signal Processing) April 1987, >> >> I found the 4/87 issue (and the rest of 1987) , but not this article. >> Are you certain of this reference? Thanks... >> >I am not sure if it was April (I believe it was), but the whole journal is >devoted to the subject of Neural Nets for that issue, and definitely exists. I went to the library today to check, and indeed the reference is correct. They are small one article publications - this is the only article in it. The cover is red with some golgi cells above a four layer net, and is subtitled "Computing with neural nets". (Ron - you said you got the rest of 1987 - since that can't be the case, do you have april 1986?) --mike
mkh1@ur-tut.UUCP (Manoj Khare) (09/24/87)
In article <1241@uccba.UUCP> finegan@uccba.UUCP (Mike Finegan) writes: >In article <759@ucdavis.UUCP>, g451252772ea@ucdavis.UUCP (g451252772ea) writes: >> > IEEE ASSP (Acoustics, Speech, and Signal Processing) April 1987, >> >> I found the 4/87 issue (and the rest of 1987) , but not this article. >> Are you certain of this reference? Thanks... >> >I am not sure if it was April (I believe it was), but the whole journal is >devoted to the subject of Neural Nets for that issue, and definitely exists. > - Mike Finegan > ...!{hal|pyramid}!uccba!finegan The article "An Introduction to Computing with Neural Nets" by Richard P. Lippmann appeared in IEEE ASSP magazine april 1987, pp 4-22. Q. Does anybody have any idea if the book "Analog VLSI and Neural Systems" by Carver A. Mead is published yet? OR Is there any way I could get his lecture notes on the related course at CalTech? Thanks in advance. ..... Manoj Khaate
mendozag@pur-ee.UUCP (09/24/87)
In article <1241@uccba.UUCP> finegan@uccba.UUCP (Mike Finegan) writes: >In article <759@ucdavis.UUCP>, g451252772ea@ucdavis.UUCP (g451252772ea) writes: >> > IEEE ASSP (Acoustics, Speech, and Signal Processing) April 1987, >> >> I found the 4/87 issue (and the rest of 1987) , but not this article. >> Are you certain of this reference? Thanks... >> >I am not sure if it was April (I believe it was), but the whole journal is ^^^^^^^ Make sure you get the magazine, tough. VMG
dlee@tut.UUCP (09/25/87)
In article <1241@uccba.UUCP> finegan@uccba.UUCP (Mike Finegan) writes: >In article <759@ucdavis.UUCP>, g451252772ea@ucdavis.UUCP (g451252772ea) writes: >> > IEEE ASSP (Acoustics, Speech, and Signal Processing) April 1987, >> >> I found the 4/87 issue (and the rest of 1987) , but not this article. >> Are you certain of this reference? Thanks... >> >I am not sure if it was April (I believe it was), but the whole journal is >devoted to the subject of Neural Nets for that issue, and definitely exists. Yes, the paper appeared in IEEE ASSP magazine, Apr. 1987. Be sure you are looking at ASSP magazine, not Journal of ASSP; they are two different publications. - Dik Lee Dept. CIS, Ohio State Univ.