First of all, thank you for your great contribution. This is really an amazing result.
I need to use adversarial training (I just touched on this issue ) in my research topic. I want to do a robust binary classification problem using logistic regression, but I am not very sure how to rewrite the loss function. For example, how to customize the loss function in the picture .Looking forward to your reply. Thank you very much.
First of all, thank you for your great contribution. This is really an amazing result. I need to use adversarial training (I just touched on this issue ) in my research topic. I want to do a robust binary classification problem using logistic regression, but I am not very sure how to rewrite the loss function. For example, how to customize the loss function in the picture .Looking forward to your reply. Thank you very much.