Adversarial Learning for a regression problem

Hi everyone!

I have studied the example on Mathworks "Train Image Classification Network Robust to Adversarial Examples". In my specific case, I'm facing not a classification problem, but a regression problem in which every image of the dataset is associated to a numeric value. So basically I am wondering how to solve this problem, how to apply adversarial learning to my regression problem. I have created the XTrain,the YTrain and the regression CNN for my dataset, but of course, since it's not a classification problem, I have no classes. Any advice to perform A.L.?

Thank you very much.

Answer 1

[Ranjeet]

Hi Daniele,

From your description, I understand that you are trying to achieve Image regression. In the example script that you pointed to “Train Image classification network robust to adversarial examples”, I suggest you to modify the CNN network by removing Softmax layer and add a FullyConnectedLayer with n inputs and single output so that you get a number as output from your network.

Also, depending upon the range on training labels (the numbers associated with an image), if the output is expected to be between 0 and 1, you may use sigmoid function as final layer. If output is expected to be between 0 and any number N, you may use max(0, N) which is ReLu.

Also, do refer the following answer to have training labels as categorical input-

https://www.mathworks.com/matlabcentral/answers/605546-image-to-number-prediction-using-cnn-regression

I assume that you already have adversarial examples in your dataset, the trained network should be robust to adversarial examples.