CMPUT 267: Machine Learning I

Course description

This course focuses on the mathematical foundations and core concepts needed for machine learning. Topics include probability, linear algebra, linear regression, classification, optimization, and model evaluation. The course also gives a light introduction to deep learning and language models and briefly examines automatic differentiation as part of the foundation of machine-learning systems engineering.

Lectures

Lectures are held in person on Tuesdays and Thursdays, 12:30–1:50 p.m., in CCIS 1-440. Lectures are not recorded. One or two lectures may move to Zoom.

Slides and course notes will be posted after class. The instructor may occasionally use the whiteboard; all whiteboard content will be included in the posted slides or course notes.

Course team and office hours

Instructor: Xi Ye
Office hours: Tuesday, 2:00–3:00 p.m., UComm 7-253

TA office hours begin September 7. See the TA office-hour schedule for times and locations.

Course work

Grading structure

• Assignments: 10%
• Midterm 1: 27.5%
• Midterm 2: 27.5%
• Final exam: 35%

Assignments

Assignments include both written and coding work and are graded on a completion basis. Late assignments are not accepted.

Midterms

There will be two 70-minute, in-class midterms, worth 27.5% each:

• Midterm 1: Tuesday, September 29
• Midterm 2: Thursday, November 5

There are no make-up midterms. If you are granted an excused absence from a midterm, its weight will be transferred to the final exam.

Final exam

The final-exam date will be announced.

AI policy

You may use AI assistants to understand concepts in this course. Treat AI as another tool, like web search, that can supplement your understanding of course content.

AI must not replace your own understanding and thinking. Do not have AI solve an assignment, do not submit AI-generated writing or code, and do not treat AI output as automatically correct.

Communication and support

Use Campuswire for all course communication. The course team will not monitor or respond to course-related email.

Post a Campuswire Q&A for questions that may help others and do not reveal assignment solutions; these posts may be anonymous. For questions specific to you, post a private Q&A visible only to the teaching assistants and instructor. For course-related issues such as a missed exam, direct-message the instructor on Campuswire.