EE 381V Statistical Machine Learning

Spring 2026

Lectures

MW 1:30-3:00pm, ECJ 1.318

Teaching assistant

TBD

References

• Primary source S. Shalev-Shwartz and S. Ben-David, Understanding Machine Learning, Cambridge University Press, 2014.
• Additional references T. Hastie, R. Tibshirani and J. Friedman, The Elements of Statistical Learning, Springer, 2009, 2nd edition.
• K. P. Murphy, Probabilistic Machine Learning: Introduction and Probabilistic Machine Learning: Advanced Topics, MIT Press, 2022-23.

Grading

Tentative grading breakdown.

• Homeworks: 30%
• Midterm exam: 30%
• Final project: 40% (project info page)

Homework policy

Homeworks are to be submitted at the beginning of the class when they are due. You may discuss homework problems with other students, but must submit your own independent solution. Late homework assignments will not be accepted.

Prerequisites

EE381J Probability and Stochastic Processes I, or equivalent, and background in linear algebra, e.g., at the level of EE381K Convex Optimization, are required.

Course description

Introduction to the theoretical underpinnings of machine learning and the derivations that transform those foundations into practical algorithms. Topics include empirical risk minimization and the probably approximately correct (PAC) learning framework; the bias-complexity tradeoff; the Vapnik-Chervonenkis dimension; Rademacher complexities; PAC-Bayes; linear predictors; boosting; model selection and validation; convex learning problems; support vector machines and kernel methods; decision trees; clustering; dimensionality reduction; generative models; and variational Bayesian methods.

Course outline

Tentative outline.

• Introduction and Foundations
  • the statistical learning framework; empirical risk minimization
  • probably approximately correct (PAC) learning; finite vs. infinite-size classes
  • the bias-complexity tradeoff
  • the Vapnik-Chervonenkis (VC) dimension
  • Rademacher complexities
• From Theory to Algorithms
  • linear prediction; boosting
  • model selection and validation
  • convex learning problems; regularization and stability
  • support vector machines; kernel methods
  • decision trees; nearest neighbors
• Beyond the PAC Learning Model
  • clustering
  • dimensionality reduction
  • generative models
  • variational Bayes methods
  • feature selection and generation
• Connections to Modern Machine Learning Practice
  • optimization and generalization in deep neural networks
  • representation learning and self-supervised contrastive methods
  • attention mechanisms and transformer architectures
  • diffusion-style generative models

Notice for students with disabilities

Students with disabilities may request appropriate academic accommodations from the Division of Diversity and Community Engagement, Services for Students with Disabilities, 471-6259, http://www.utexas.edu/diversity/ddce/ssd/.

Emergency instructions

All occupants of university buildings are required to evacuate a building when a fire alarm and/or an official announcement is made indicating a potentially dangerous situation within the building. Familiarize yourself with all exit doors of each classroom and building you may occupy. Remember that the nearest exit door may not be the one you used when entering the building. If you require assistance in evacuation, inform your instructor in writing during the first week of class.

For evacuation in your classroom or building:

1. Follow the instructions of faculty and teaching staff.
2. Exit in an orderly fashion and assemble outside.
3. Do not re-enter a building unless given instructions by emergency personnel.