Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > alt.language.artificial > #9

Re: Preceptors and wieght or Bias

From ruben safir <ruben@mrbrklyn.com>
Newsgroups alt.language.artificial
Subject Re: Preceptors and wieght or Bias
Date 2017-03-27 21:51 -0400
Organization PANIX Public Access Internet and UNIX, NYC
Message-ID <obcfiq$510$2@reader1.panix.com> (permalink)
References <obcff3$510$1@reader1.panix.com>

Show all headers | View raw


On 03/27/2017 09:49 PM, ruben safir wrote:
> When I have a preceptor, shouldn't I have a weighted bias that is added
> to the system?  How is that done.  I'm reading this
> 
> https://liucs.net/cs691s17/perceptrons.html
> 
> and trying to coralate it with pg 728 of Stuart J. Russell and Peter
> Norvig - 3rd edition
> 

To quote the text:

Just as in linear regression models, each unit has a dummy input a0 = 1
with an associated weight w0,j .

Back to alt.language.artificial | Previous | NextPrevious in thread | Find similar | Unroll thread


Thread

Preceptors and wieght or Bias ruben safir <ruben@mrbrklyn.com> - 2017-03-27 21:49 -0400
  Re: Preceptors and wieght or Bias ruben safir <ruben@mrbrklyn.com> - 2017-03-27 21:51 -0400

csiph-web