Implementation of FGSM, I-FGSM, MI-FGSM 

Adversarial attacks became very popular since first attack on SVM model. Then came first white box attack known as FGSM (Fast Gradient Sign Method) paper link then on top of this in span on 2 years as an enhancement came Iterative FGSM (I-FGSM) paper link and then again in span of 2 years new enhancement cam with adding Momentum boosting these adversarial attack known as Momentum Iterative Fast Gradient Sign Method(MI-FGSM) paper link. 

As part of my bigger project I studied these method as literature survey and implemented these methods in PyTorch and have provided the implementation of the same in my GitHub repo.