Crash course on Deep Learning

This lecture provides a crash course on deep learning, starting with the Mark I Perceptron, the first implementation of the perceptron algorithm. It covers neural networks inspired by neuroscience, the Universal Approximation Theorem, gradient descent optimization, and different algorithms like SGD, Momentum, and ADAM. The lecture delves into the practical aspects of training neural networks, including mini-batch gradient descent, and the use of Pytorch for optimization. It concludes with a preview of upcoming topics such as convolutional neural networks and the contributions of prominent figures like Yoshua Bengio, Geoffrey Hinton, and Yann LeCun.