ECE364 focuses on auto-differentiation tools like PyTorch used with basic machine learning algorithms (linear regression, logistic regression, deep nets, k-means clustering) and extensions in custom methods to fit specific needs. Auto-differentiation tools are essential for data analysis, and a solid understanding of them is increasingly important in many disciplines. In contrast to existing courses focusing on algorithmic and theoretical aspects of Machine Learning, the focus here is on implementation with auto-diff tools.
And that's a wrap. This course was offered for the first (and currently last) time in Spring 2025. I'm leaving the content up in case anyone else finds it useful, and I'll keep adding to it here and there as I continue refining my own understanding of some topics. I had a great time teaching this course, but before wrapping up, I want to give a huge shoutout to my amazing course staff: Neeraj Gangwar, Suyuan Wang, Vishesh Prasad, and Jeff Shin. I know I’m the professor and my name's always at the top, but honestly, none of this would've been possible without them—I owe them a ton. Thanks again to my staff and students for an incredible semester, and don’t be afraid to ping!
News
-
January 21, 2025
Welcome to the ECE364 website for the Spring 2025 semester. The site is continuously evolving and will be updated regularly throughout the semester, so be sure to check back often for the latest information. I’m sure there are many typos and such. If you see any errors, please submit an issue in the website’s github repo! Raising issues improves the course site, ups your GitHub participation stats, and will be rewarded in other ways as well!
Schedule Quicklinks
People
Alumni graduate teaching assistants
Past graduate teaching assistants. Most have created/contributed to the notes and problems/solutions.
Neeraj Gangwar
Suyuan Wang
Alumni Undergraduate Teaching Assistants
Many of our past CAs. Most have moved on to successful careers but their time grading and hosting office hours won't be forgotten.
Vishesh Prasad
