3 ms·
Disclaimer: Googler here, but not on the TF team and opinions are my own. Tangential, but related. For the Deep learning folks I recommend having a look at ten
by marcyb5st 4y ago
Disclaimer: Googler here, but not on the TF team and opinions are my own.
Tangential, but related. For the Deep learning folks I recommend having a look at tensorflow probability or some pytorch analogue. For instance, following [1] you see that bolting the capability to model at least aleatoric uncertainty is super simple.
The benefits are enormous. In fact, apart from technical aspects that such an approach can unlock (eg if the output is favorable but the uncertainty is too high still don't act on the predictions because the cost of a acting on a false positive is >> than non acting on a false negative) I found that being able to output the model uncertainty goes a long way in convincing non technical stakeholders about trusting predictions.
[1] https://blog.tensorflow.org/2019/03/regression-with-probabilistic-layers-in.html https://blog.tensorflow.org/2019/03/regression-with-probabil...
- anonymousDan 4y agoYou might find the theory of conformal prediction interesting given your comment: https://arxiv.org/abs/0706.3188 https://arxiv.org/abs/0706.3188
- nerdponx 4y agoI didn't know there was a name for this and established set of tools for it. Good things come from even the trashiest of HN front page articles, thank you. Edit: wow, I don't think I've seen "support vector machine" mentioned even once anywhere in the last 2 years. Times have changed! I wonder how this toolbox has been updated in the age of deep learning and probabilistic programming.
- anonymousDan 4y agoI think this is supposed to be a general technique that can be applied to any classification algorithm. Whether the performance is practical for DNNs is another matter. As I understand it there is ongoing research in the area. I think there is overlap between the inventors of SVMs and conformal prediction, so that might explain why they are mentioned!