Deep Neural Networks and Applications with Tensorflow

This course will introduce deep neural networks, the main kinds of architectures, explore some applications, and use Python and Tensorflow 2.0. The course will assume some familiarity with programming in Python, probability, and statistics, linear algebra, and calculus. It will also assume some familiarity with machine learning and/or artificial intelligence although this material will be briefly reviewed. Tentative topics will include • Review of Machine Learning with Python, Pandas, Sklearn and Tensorflow 2.0 • Multi-layer Neural Networks • Convolutional Neural Networks • Sequence models (eg. Recurrent Neural Networks) • Generative Models Assessment will be based on homework exercises, student-developed tutorials, a midterm and a group project. Ph.D. students will also be expected to develop a 10-minute video reviewing a paper with novel code applying the technique.