(Same as Computer Science M146.) Lecture, four hours; discussion, two hours; outside study, six hours. Requisites: course 131A or Civil and Environmental Engineering 110 or Mathematics 170A or 170E or Statistics 100A; Computer Science 32 or Program in Computing 10C; Mathematics 33A. Introduction to breadth of data science. Foundations for modeling data sources, principles of operation of common tools for data analysis, and application of tools and models to data gathering and analysis. Topics include statistical foundations, regression, classification, kernel methods, clustering, expectation maximization, principal component analysis, decision theory, reinforcement learning and deep learning. Letter grading.

Review Summary

Clarity
10.0 / 10
Organization
10.0 / 10
Time
5-10 hrs/week
Overall
10.0 / 10

Reviews

    Quarter Taken: Spring 2023 In-Person
    Grade: A+

    Suhas Diggavi is very bright and truly an accomplished computer engineer. His lectures are very clear, but can get a little too deep into confusing notation. However, this notation is what's used in machine learning literature, so I can't exactly fault him for using it—does prepare us for actual ML readings well.

    The course taught cool concepts, albeit a little slowly. I spoke with another ML professor (Jonathan Kao) about the course content, and he says that some of it is not very useful (KNN). Not really an indictment on the professor, but more so on the curriculum.

    What's great is how small the class was. Only 19 students are enrolled, and only about 7 attend lecture. Very interactive and immersive learning experience.

    Rida is a great TA. Very knowledgeable, fills in gaps that were present in lecture, and always is willing to expand on concepts.

    Similar to lecture, not many people attend discussion (only 3 including myself). This makes it more like an office hour where we can interact with the TA very well. Would 100% recommend Rida.

Course

Instructor
Suhas N. Diggavi
Previously taught
24S 23S 22S 18S
Formerly offered as
EL ENGR M146

Grading Information

  • No group projects

  • Attendance not required

  • 1 midterm

  • Finals week final

  • 100% recommend the textbook

Previous Grades

Grade distributions not available.