Teaching
Fall 2024
ECE 381K 3 - Applied Machine Learning
:
Unique 17685, TuTh 12:30-2pm. ECJ 1.214.
Syllabus
This course is strictly for graduate students. ECE undergrads should take EE461P instead.
One cannot get credit for both EE461P and EE380L due to the high degee of overlap on the applied ML aspects.
Also, due to anticipated demand, pre-enrollment is restricted to ECE graduate students till the first day of class, when
other qualified students are welcome to enroll.
Pre-reqs are stated in the course descriptor.
If the class is full but you are really interested, do persist and be on the wait list; invarially adequate number of spots open after the first week of classes.
MIS 382N: ADVANCED MACHINE LEARNING - MSBA (Unique 04685/90):
MW 2pm and 3:30pm, GSB3.130.
This is not a regular course but an "Option III" course. Hence all sections of this course are restricted to students in
the McCombs "Masters in Business Analytics" (MSBA) Programs. Other UT students (including ECE students) cannot be enrolled for this course.
Spring 2024
EE461P Data Science Principles
TuTh 11am-12:30pm. (17170, EER 1.518)
This course is meant for senior/advanced-junior undergrads in ECE with adequate background in math/stats and programming
(see pre-requisites). Graduate students are not allowed; they should instead consider ECE 381K-3: Applied Machine Learning.
MIS 382N: ADVANCED MACHINE LEARNING - MSBA (Unique 04610):
This course is restricted to students enrolled in the online MSBA program with McCombs.
Previously offered Courses include
EE 381V Fair Transparent Machine Learning. (Fall 2022) :
This is a
seminar-oriented special topics course in machine learning and an updated version of the
Course
I taught in Spring 2019 (see below), with emphasis on fairness and explainability of machine learning based solutions. Like that course, a pre-requisite is that you must have already taken one graduate level course in machine learning/data science.
So not a suitable course to take if you want to pick up basics of machine learning, in fact you will need to drop for not satisfying the pre-requisite. CS graduate students can enroll once the semester starts without needing my permission.
Also exceptions can be made for PhD students from other disciplines (e.g. law) who are actively working in AI ethics and have adequate statistical background - such students are requested to email me if they are interested.
Course includes theory, algorithms, statistics and social perspectives.
Reading List (use your @utexas email to access)
EE381V FAT ML: Spl Topics in Machine Learning (Spring 2019)
Pre-reqs: At least one graduate course completed in Data Mining/Machine Learning. Online courses do not count.
This is an advanced, seminar-oriented course. We shall study recently published papers relevant to the development of responsible and trustworthy data driven automated decision systems. Solid background in pattern recognition/machine learning is assumed. Key topics include building explainable ML models, black-box explainability, algorithmic fairness, adversarial ML, robust statistical modeling, and privacy aware data mining. Coursework will mainly involve paper presentations, critiques and discussion, a mini coding-based project and a major term project on developing some aspects of a responsible ML system.
Course READING LIST
This edition of the course focussed on Big Data Analytics for Healthcare.