# Is classifier able to say there's no-such-case?

I am a starter in ML and I need some help...

## The problem

Assume that I have a classifier which can classify left hand / right hand well. I am curious whether it can decide whether there's a hand in the image?

Because if it is not the case, I have to collect additional data that are labeled as "no-hand". But there's so many things that are "no-hand" such like dogs, cats, flies... I don't think it is possible to collect all those possible "no-hand" data..

## My intuition

Would it be reasonable that we use some activation functions like [sigmoid] as the final outputs for the classifier(neural network) and we consider the outputs for both left/right classes as the scores that classifier think how far the input is from the left/right hands classes.

Then if the outputs are below some threshold for both left hand and right hand we can say there's no hand in this image input.

• Welcome to the site! Since the classifier says L or R , means that it should definitely fall under either of those classes. 1st how does a classifier works, it will classify what ever data you give either as L or R. It doesn't matter what you give it would it can only tell you either R or L. Now coming to your question you cannot do that with the existing classifier as it was not trained to tell you Yes or No. Even if you give some threshold also I don't think that the inference would be right. Jan 12, 2018 at 7:23
• With a bit of creativity this problem can indeed be solved. Jan 12, 2018 at 7:27

Taking your questions one after the other:

Assume that I have a classifier which can classify left hand / right hand well. I am curious whether it can decide whether there's a hand in the image?

No. It cannot. A model which is trained on hand/no-hand can do that. Not this. At most, it shall predict one of the either (left/hand) with a very low confidence score. But, it definitely is not a nice idea to use it for a hand/no-hand identification.

But there's so many things that are "no-hand" such like dogs, cats, flies... I don't think it is possible to collect all those possible "no-hand" data

You are trying to do build a binary hand.no-hand classification model, not a hand/cat/dog/... style multi-label classification model. So, you just need to have a bunch of images with a hand in it, and some more which doesn't have one.

Would it be reasonable that we use some activation functions like [sigmoid] as the final outputs for the classifier(neural network) and we consider the outputs for both left/right classes as the scores that classifier think how far the input is from the left/right hands classes.

Many classifiers do not directly give you L or R. They will give you the option which has a higher decision metric. For Naive Bayes this would be the class with the higher probability between $p(L|x)$ and $p(R|x)$. This is true for most classifiers.
For example, if $p(L|x)=0.01$ and $p(R|x)=0.05$ then we can reject both hypotheses and say the novel example $x$ belongs to neither class.
You should look further into anomaly detection algorithms, you can also use p-value through the GLRT to test if your prediction falls enough within the class such that it can be classified as such. For example, set you p-value, $p = 0.05$, if the p-value of a given example falls below this threshold then do not attribute a class to it and flag it as anomalous.