Re:Making a neural network say “I Don’t Know”: Bayesian NNs ...
by JD557 on Wednesday November 28, 2018 @11:34AM
(#18552900)
My Master thesis supervisor had a fun trick for this, that can be used not only on on NNs, but also on other classifiers: https://arxiv.org/pdf/1011.3177.pdf
The gist is, you train a single classifier that will work as two classifiers (with certain restrictions): one that says "false/not-false" and another that says "true/not-true".
If the output of the classifier is "not-false" and "not-true", then you consider that as a "I don't know".
Re:Making a neural network say “I Don’t Know”: Bayesian NNs ...
by yboris on Wednesday November 28, 2018 @02:30PM
(#18554806)
Is this what you mean:
For example, in MNIST, your model would output 20 probabilities: (1-yes: 0.8, 1-no: 0.2, 2-yes: 0.1, 2-no: 0.9, etc) where each pair sums to 1?
The gist is, you train a single classifier that will work as two classifiers (with certain restrictions): one that says "false/not-false" and another that says "true/not-true".
If the output of the classifier is "not-false" and "not-true", then you consider that as a "I don't know".