Neural Network Linear Regression

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Neural Network Linear Regression

Jacques.Grobler
Hi all

I am very much new to Machine learning and Neural Networking

I have a question: Can you apply Linear Regression (or any non Logistic
Regression) in a Neural Network setup?

I have searched the internet and can simply not find any examples of where
Linear Regression is applied.

Any help will be appreciated

Thank you



--
Sent from: http://octave.1599824.n4.nabble.com/Octave-General-f1599825.html

_______________________________________________
Help-octave mailing list
[hidden email]
https://lists.gnu.org/mailman/listinfo/help-octave
Reply | Threaded
Open this post in threaded view
|

Re: Neural Network Linear Regression

Doug Stewart-4


On Sat, Sep 16, 2017 at 12:09 PM, Jacques.Grobler <[hidden email]> wrote:
Hi all

I am very much new to Machine learning and Neural Networking

I have a question: Can you apply Linear Regression (or any non Logistic
Regression) in a Neural Network setup?

I have searched the internet and can simply not find any examples of where
Linear Regression is applied.

Any help will be appreciated

Thank you




What would be the reason of applying it.

The short answer is no. the long answer is that you can invent many different ways of doing NNs. But is in the normal ones I have not seen Linear regression used.  

_______________________________________________
Help-octave mailing list
[hidden email]
https://lists.gnu.org/mailman/listinfo/help-octave
Reply | Threaded
Open this post in threaded view
|

Re: Neural Network Linear Regression

Jacques.Grobler
Hi Doug

I have developed a program that inputs the minimum requirements for the
design of mechanical equipment, the program then outputs a 3D model with
bill of materials with masses, qties, areas etc for estimating purposes.
However the time laps is relatively long and computationally expensive.

So I am investigating the use of Machine Learning in order to determine the
outputs faster and less expensive.

I have managed to do a representation using simple linear regression however
I have only managed to output a single variable at a time, so I have to run
multiple training sets in order to get multiple outputs and I believe that
with NN I would be able to do a single learning algorithm to determine
multiple outputs.

At this stage any type of help would be greatly appreciated as I have no
idea of which direction to go



--
Sent from: http://octave.1599824.n4.nabble.com/Octave-General-f1599825.html

_______________________________________________
Help-octave mailing list
[hidden email]
https://lists.gnu.org/mailman/listinfo/help-octave