Quiz: Training GANs to Break Different Models

In this chapter, we looked into the fundamentals of adversarial examples and how to attack and confuse a CNN model with FGSM (Fast Gradient Sign Method). We also learned how to train an ensemble classifier with several pretrained CNN models via transfer learning on Kaggle’s cats vs. dogs dataset, following which, we also learned how to use an accimage library to speed up image loading even more and trained a GAN model to generate adversarial examples and fool the image classifier.

Test your understanding

You can attempt the quiz below to test your understanding: