Deep Learning

3 credits · Graduate

Deep Learning is a subfield of Machine Learning; in this course, we study concepts that build on the fundamentals of neural networks and machine learning. This extension of concepts may include topics such as variational autoencoders, dilated convolutional networks, generative adversarial networks, adversarial examples, attention mechanisms, the transformer architecture, language models such as ELMo and BERT. In doing so, we improve our understanding of how the more basic systems work, and explore foundations such as optimization techniques. CALENDAR NOTES: Students are expected to have a knowledge of fundamental concepts of Machine Learning. Students are also expected to strong mathematical skills in multivariate calculus, linear algebra (e.g. eigendecomposition), and probability (e.g. multi-dimensional Gaussian pdf).