AIbase
Product LibraryTool Navigation

MSAP

Public

In this work, we extend the FGSM method proposing multistep adversarial perturbation (MSAP) procedures to study the recommenders’ robustness under powerful methods. Letting fixed the perturbation magnitude, we illustrate that MSAP is much more harmful than FGSM in corrupting the recommendation performance of BPR-MF.

Creat2020-05-26T17:27:37
Update2024-06-18T20:02:54
2
Stars
0
Stars Increase

Related projects