The Situation: Our Support Vector Machine (SVM) and Convolutional Neural Network (CNN) both did pretty well on MNIST.
The Illusion: MNIST is considered a "solved" problem. It's too easy. Black and white, perfectly centered digits don't represent the complexity of the real world.
The Goal: We need to test our models on a much harder dataset to truly see why Deep Learning revolutionized AI and surpassed classic ML techniques.
