| layout | nav_order | title | description |
|---|---|---|---|
page |
999 |
FAQs |
Frequently Asked Questions |
Q: What are the prerequisites for this course?
A: You need to have completed either:
- A course on ML algorithms (e.g., CSE 151) is absolutely necessary.
- A course on either database systems internals (e.g., CSE 132C) or operating systems (e.g., CSE 120) is also necessary.
- DSC 102 suffices as a perequisite for both of the above aspects.
- Substantial project or industrial experience on relevant topics can be substituted for prior coursework, subject to the instructor's consent. Email the instructor if you would like to enroll but are unsure if you satisfy the prerequisites.
Q: I have relevant industry experience but haven't taken the prerequisite courses. Can I still enroll?
A: Yes, substantial project or industrial experience can be considered as a substitute for prerequisites. Email the instructor directly to discuss your background and eligibility.
Q: What is the main focus of this course?
A: This is a research-based course that explores systems for machine learning, combining elements of ML/AI and systems. In this year, we will focus more on LLMs, its related system support, optimizations, deployment challenges, and emerging problems.
Q: Who is this course designed for?
A: The course is primarily tailored for:
- MS students
- PhD students
- Advanced undergraduate students who are interested in systems for scalable data science and ML engineering.
Q: What are the suggested readings for this course?
A: Suggested Textbooks
- Recommended: Data Management in Machine Learning Systems, by Matthias Boehm, Arun Kumar, and Jun Yang (Free ebook via UCSD VPN).
- Additional (optional) for background/foundations on the respective component areas:
- Machine Learning, by Tom Mitchell (McGraw Hill).
- Deep Learning, by Ian Goodfellow, Yoshua Bengio, and Aaron Courville (MIT Press)
- Operating Systems: Three Easy Pieces, by Remzi and Andrea Arpaci-Dusseau (Free ebook).
- mlsysbook.ai
Q: What is the best way to get help in this course?
A: Your best avenues are to go to office hours held by the course staff, or to ask questions on Piazza. Course staff will be monitoring Piazza frequently and will try to answer your question quickly and thoroughly.
Q: Where will our grades for assignments be displayed for the course?
A: Grades will be displayed on Gradescope for the written and autograded portions for all assignments (homeworks, reading summaries, and exams).
Q: I noticed a mistake in the grading of the written portion of my homework. How can I get this fixed?
A: To get this fixed, you must submit a regrade request via Gradescope before the regrade deadline. This is known as the regrade request window. We unfortunately will not accept any regrades after the window has closed. All regrade deadline dates are posted on the same Ed post that releases the assignment grades and solutions.
Q: This is my first time being a scribe. I'm not too sure what my role is.
A: First off, we're glad that you're taking some time to understand our expectations! The expectations from a scribe are simple: the scribe should faithfully capture the professor's explanations in the lecture. We expect the notes to have the same structure/main sections as the slides, and many similar wordings as well. Additionally, the notes should also have helpful examples given by the professor in the class, additional context for different topics when needed, etc. The TAs will assess your notes based on the contents in the slides.
You should not, however, try to rearrange the content as you see fit, or add in material from the additional readings provided. The scribe notes do not have to be a self-contained document on the topic - we're only looking for you to capture what was taught in class.
<script src="../assets/darkmode.js"></script> <script> window.addEventListener("DOMContentLoaded", (event) => { onLoad(); }); </script>