An introductory deep learning course, which emphasizes the fundamental algorithmic advances that have made modern deep learning possible, including forward- and reverse-mode automatic differentiation and stochastic optimization. The course will also cover standard deep learning architectures such as convolutional networks, recurrent networks, and transformers, and their applications to computer vision, natural language processing, speech processing, and reinforcement learning. Recommended prerequisites: undergraduate-level multivariable calculus, linear algebra, probability, and machine learning, and comfort with programming in python.
