PUBH8475

This course surveys emerging topics at the intersection of machine learning, data analytics, and big data. We will introduce statistical and computational techniques for both prediction and inference, with applications to business analytics, engineering, and biomedical research. Students should have prior exposure to basic machine learning and data mining (e.g., PubH 7475, STAT 8053, or an equivalent course). Topics include data exploration; optimization for learning; high-dimensional methods; deep learning (FNN, CNN, RNN/LSTM) and modern advances (Transformers, diffusion models); recommender systems; graphical models; text mining and language models; and causal machine learning. Learning and data mining (e.g., Pubh7475, Stat8053, or a similar course). Topics include the following:

Data exploration and data science; Optimization for machine learning; High-dimensional analysis: Prediction and inference; Deep neural network learning: basics (FNN, CNN, RNN/LSTM); Advanced topics (Transformers, Diffusion models, etc);
Recommender systems: personalized prediction; Undirected and directed graphical models; Unstructured data and text mining: Numerical embedding and language models; Causal Machine Learning.

prereq: Pubh7475 or Stat8053 or a similar course, or permission of instructor; familiarity with programming in R or Python.

Credit will not be granted if credit has been received for: STAT 8056

002223

01358

Spring Even Year