1

I found some similar questions (such as How best to deal with "None of the above" in Image Classification?) but nothing very recent.

In my problem I have 10 trained classes of interest, but occasionally I get a garbage image of some kind, and of course the classifier just guesses from the options it has. Sometimes it spreads the softmax probability out, but often it is fairly confident in one of its choices. This is not unreasonable to me since it lives in a world of "given an image of one of 10 things, which is it?", but it is problematic for a system I intend to integrate into deliverable software.

A "none of the above" class would be okay, but finding training examples for such a thing is a little dicey.

What approaches are effective for this problem?

smci
  • 32,567
  • 20
  • 113
  • 146
Mastiff
  • 2,101
  • 2
  • 23
  • 38
  • 2
    The standard practice is to create one or more extra 'junk classes'/'garbage classes', and add some labeled images to the training set, so the model correctly classifies them as garbage, not as one of your 10 classes-of-interest. Read about ["junk class" or "garbage class" on the sister site CrossValidated](https://stats.stackexchange.com/search?q=garbage+class) – smci Oct 18 '19 at 01:18

0 Answers0