There has been a clear trend that people enjoy using deep learning methods to solve problems in the computer vision field. Has one of your classmates or colleagues ever shown off their latest image classifier to you? Now, with GANs, you may actually get the chance to show them what you can do by generating adversarial examples to break their previous models.
We will be looking into the fundamentals of adversarial examples and how to attack and confuse a CNN model with FGSM (Fast Gradient Sign Method). We will also learn how to train an ensemble classifier with several pre-trained CNN models via transfer learning on Kaggle's Cats vs. Dogs dataset, following which, we will learn how to use an accimage library to speed up your image loading even more and train a GAN model to generate adversarial examples and fool the image classifier...
