About this Course

Data, Inference, and Decisions

This course develops the probabilistic foundations of inference in data science. It builds a comprehensive view of the decision-making and modeling life cycle in data science, including its human, social, and ethical implications. Topics include: frequentist and Bayesian decision-making, permutation testing, false discovery rate, probabilistic interpretations of models, Bayesian hierarchical models, basics of experimental design, confidence intervals, causal inference, robustness, Thompson sampling, optimal control, Q-learning, differential privacy, fairness in classification, recommendation systems and an introduction to machine learning tools including decision trees, neural networks and ensemble methods.

This class is listed as Data 102.

Course email address

data102@berkeley.edu

If you are requesting extensions or if you have personal concerns, please contact us via the email address above.

Announcements

All course announcements will be made on Ed.

Important Information

NOTE: The below information reflects our plans for the bulk of the semester. Note that all activities will be held remotely for the first two weeks, per campus. However, once campus allows, we plan to return to primarily in-person lab, discussion, and office hours, although we do plan to have remote options available for all class activity beyond the first two weeks. Lecture will be held remotely throughout the entire semester.

Lecture: Tuesdays and Thursdays from 9:30 AM to 11:00 AM, remotely on Zoom. See Ed posts (link coming soon) for the zoom link. Lecture videos will be recorded and links will be available on the course website within a few hours after lecture.

Discussion session: Wednesdays in-person (with virtual sessions for those unable to be on campus: more info TBA). See Ed posts (link coming soon) for more information on this. Attendance is highly encouraged but not mandatory.

Lab: Mondays in-person (with virtual sessions for those unable to be on campus: more info TBA). You can complete lab assignments on your own time, but you are highly encouraged to attend lab sessions to work with your classmates and get help from the staff.

Contacting Course Staff: The best way to contact course staff is using Ed, using a private post as needed—we’ll be checking it regularly and should respond to most questions within a day or less. If you need to reach out to course staff and Ed isn’t suitable, you can email [data102@berkeley.edu], and you should get a response within a few days. Please avoid emailing professors or GSIs directly!

Office Hour, Lab, and Discussion Schedule

Please see Ed posts for Zoom links for remote events.

Prerequisites

While we are working to make this class widely accessible we currently require the following (or equivalent) prerequisites :

Principles and Techniques of Data Science: DS100 covers important computational and statistical skills that will be necessary for DS102.
Probability: Probability and Random Processes EECS126, or Concepts of Probability STAT134, or Probability for Data Science STAT140, or Probability and Risk Analysis for Engineers IEOR172. EECS126 and STAT140 are prefered. These courses cover the probabilistic tools that will form the underpinning for the concepts covered in DS102.
Math: Linear Algebra & Differential Equations Math54, or Linear Algebra MATH110, or both Designing Information Devices and Systems I EE16A and Designing Information Devices and Systems II EE16B, or Linear Algebra for Data Science Stat89a, or Introduction to Mathematical Physics PHYSICS89. We will need some basic concepts like linear operators, eigenvectors, derivatives, and integrals to enable statistical inference and derive new prediction algorithms.