home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.neural-nets
- Path: sparky!uunet!utcsri!torn!csd.unb.ca!morgan.ucs.mun.ca!raghu
- From: raghu@morgan.ucs.mun.ca (Raghu B)
- Subject: Back Propogation - ??????
- Message-ID: <1992Jul28.170531.3687@morgan.ucs.mun.ca>
- Keywords: Learning - Strange
- Organization: Memorial University of Newfoundland
- Date: Tue, 28 Jul 1992 17:05:31 GMT
- Lines: 28
-
- Thanx for the response for my earlier query. I found another
- strange behaviour in my bp algorithm. I have in my plot of
- actual and neural values something like this always.
-
-
- :
- : : /
- : : / :
- : : / :
- : / : : neural values
- : / / actual values
- :
- :
- :
- :
- :------------------------------------>
- time --->
-
- I always find a bowl shaped nature in the neural behaviour.
- The error is at an acceptable level. Is bowl shape a characteristic
- of bp algorithm ? I am curious to know more about it.
-
- Waiting for positive responses from dear netters.
-
- Raghu,Balasubramanian
-
-
-
-