Xref: utzoo sci.electronics:3552 comp.ai:2094 comp.ai.neural-nets:180
Path: utzoo!attcan!uunet!husc6!bbn!uwmcsd1!ig!agate!ucbvax!decwrl!labrea!glacier!jbn
From: jbn@glacier.STANFORD.EDU (John B. Nagle)
Newsgroups: sci.electronics,comp.ai,comp.ai.neural-nets
Subject: Re: Sigmoid transfer function
Message-ID: <17615@glacier.STANFORD.EDU>
Date: 6 Aug 88 16:58:05 GMT
References: <1945@aecom.YU.EDU>
Reply-To: jbn@glacier.UUCP (John B. Nagle)
Organization: Stanford University
Lines: 14


     Recognize that the transfer function in a neural network threshold unit
doesn't really have to be a sigmoid function.  It just has to look roughly
like one.  The behavior of the net is not all that sensitive to the 
exact form of that function.  It has to be continuous and monotonic, 
reasonably smooth, and rise rapidly in the middle of the working range.
The trigonometric form of the transfer function is really just a notational
convenience.

     It would be a worthwhile exercise to come up with some other forms
of transfer function with roughly the same graph, but better matched to
hardware implementation.  How do real neurons do it?

					John Nagle