Unsupervised Learning

4 credits

Broad view of the field of unsupervised learning, in particular its most common methods and use cases. Topics include descriptive statistics; clustering; projection, singular value decomposition, and spectral embedding; common probability distributions; density estimation; graphical models and latent variable modeling; sparse coding and dictionary learning; autoencoders, shallow and deep; and self-supervised learning. Prerequisites: DSC 255R. Restricted to major codes DS78 and DS79. All other students with graduate standing may be considered as space permits.