3 ms·
Adding high frequency noise "fools" ML but not the human eye. It feels like this is a general failure in regularization schemes. Why not try training multiple
by jmcminis 10y ago
Adding high frequency noise "fools" ML but not the human eye. It feels like this is a general failure in regularization schemes.
Why not try training multiple models on different levels of coarse grained data? Evaluate the image on all of them. Plot the class probability as a function of coarse graining. Ideally its some smooth function. If it's not, there may be something adversarial (or bad training) going on.
- lerid 10y agoThis seems like an obvious thing to try, but technically the convolutional structure already looks at different scales (in a way similar to mipmaps).
- jmcminis 10y agoI've built a CNN before and that's not my understanding of it. The high frequency noise changes the output of the first layer on the CNN. This is what gets pooled as you go deeper into the net. Coarse graining is like getting rid of the weights you have for the first layer and replacing them with something uniform (average the smallest details together).