Visit

Students will learn deep neural network fundamentals, including, but not limited to, feed-forward neural networks, convolutional neural networks, network architecture, optimization methods, practical issues, hardware concerns, recurrent neural networks, dataset acquisition, dataset bias, adversarial examples, current limitations of deep learning, and visualization techniques. We still study applications to problems in computer vision and to a lesser extent natural language processing and reinforcement learning. There will also be a session on understanding publications in deep learning, which is a critical skill in this fast moving area.