loss return question

[PiggyGaGa]

Hello, than you for this awesome project, which is very useful to help me understand the method in NNPU paper. But I am confused abount the NNPU loss, In the file pu_loss.py line 51, the loss output returns gamma * negative_risk. I also saw this in NNPU paper Algorithm 1 Line 9 . In the paper, author said, the loss go along with negative_risk with a step size discounted by gamma to make mini-batch less overfitted when negative_risk < -beta. I can not understand this. Besides, I am confused why there is no positive_risk when negative_risk < -beta .

Hoping for your reply, thank you.