Solved

In Neural Network training, how to represent instances that do not belong to any class

Posted on 2012-03-18
4
352 Views
Last Modified: 2013-11-25
Also, how should that be reflected into the design of the network itself?
I mean if the output of 0001 means 1 and the output of 0010 means 2, can we train the network to produce 0000 for instances that do not belong to any class?
0
Comment
Question by:Lamiaaa
  • 2
4 Comments
 
LVL 31

Expert Comment

by:GwynforWeb
ID: 37735729
Are you using a forward-feeding (ie layers proceeding left-right) or a feedback network?
0
 
LVL 35

Assisted Solution

by:mccarl
mccarl earned 500 total points
ID: 37735762
Theoretically, yes, 0000 would mean no class, but in practical terms, you would have to evaluate the performance of that. It would depend on all the usual things with NN's, such as the input data, number of layers, number of neurons, how you train it, etc.
0
 

Author Comment

by:Lamiaaa
ID: 37735810
Thanks guys,

yes, it is forward-feeding.

Are there any sampling approaches (proposed to solve similar problems) to build a training set that appropriately represent the 0000 cases?

It seems difficult because "every thing else" is a very huge and diverse set!
0
 
LVL 35

Accepted Solution

by:
mccarl earned 500 total points
ID: 37735880
It is a bit hard to say, as only you know your problem domain. One possibility might be a random sample that is the same size as the total of all 'classified' samples, ie. of the total training set, half is "everything else' and the other half is 'one of the target classes'.

I would base this on the performance of the network as it stands. So first, you would include these 'everything else' inputs into your verification set, and see how the network performs. Then depending on the results of the verification step, modify the training set to see what improvements can be gained.
0

Featured Post

Gigs: Get Your Project Delivered by an Expert

Select from freelancers specializing in everything from database administration to programming, who have proven themselves as experts in their field. Hire the best, collaborate easily, pay securely and get projects done right.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Suggested Solutions

Title # Comments Views Activity
Homework Math Question 1 104
Odds of picking games correctly 4 89
Math Question: What is the max distance of x? 5 76
rhino JavaScript import, load 25 82
How to remove superseded packages in windows w60 or w61 installation media (.wim) or online system to prevent unnecessary space. w60 means Windows Vista or Windows Server 2008. w61 means Windows 7 or Windows Server 2008 R2. There are various …
If you haven’t already, I encourage you to read the first article (http://www.experts-exchange.com/articles/18680/An-Introduction-to-R-Programming-and-R-Studio.html) in my series to gain a basic foundation of R and R Studio.  You will also find the …
The goal of the video will be to teach the user the difference and consequence of passing data by value vs passing data by reference in C++. An example of passing data by value as well as an example of passing data by reference will be be given. Bot…
The viewer will learn additional member functions of the vector class. Specifically, the capacity and swap member functions will be introduced.

785 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question