15.097 | Spring 2012 | Graduate

Prediction: Machine Learning and Statistics

Instructor Insights

Course Overview

This page focuses on the course 15.097 Prediction: Machine Learning and Statistics as it was taught by Professor Cynthia Rudin in Spring 2012.

This introductory course provides a selection of the most important topics from the subjects of machine learning and statistical methods.

Course Outcomes

Course Goals for Students

  • Gain a foundational understanding of how machine learning and statistical algorithms work
  • Explore statistical learning theory, which provides the mathematical foundation for machine learning algorithms
  • Learn about Bayesian analysis
  • Leave the course with a toolbox of algorithms that can be used on students’ own datasets

Possibilities for Further Study/Careers

15.097 prepares students for more in-depth study in fields involving predictive modeling.

Curriculum Information

Prerequisites

  • Permission of instructor.
  • Knowledge of MATLAB® may be helpful.

Offered

This course was offered once in Spring 2012.

Instructor Insights

In the following paper, Professor Cynthia Rudin describes various aspects of how she taught 15.097 Prediction: Machine Learning and Statistics. Topics include the organizational structure of the course and the strategy of using of large-font lecture notes.

Teaching “Prediction: Machine Learning and Statistics” (PDF)

Assessment

The students’ grades were based on the following activities:

  • 50% Problem sets
  • 10% Course project proposal
  • 2% Course project advertisement
  • 38% Course project paper and talk

Student Information

Enrollment

36 students

Breakdown by Year

Mostly graduate students

Breakdown by Major

Predominantly students from the Operations Research Center

Typical Student Background

Students had a general background in mathematics. Topics from machine learning and statistical methods were new to them. 

How Student Time Was Spent

During an average week, students were expected to spend 12 hours on the course, roughly divided as follows:

Lecture

  • Met 2 times per week for 1.5 hours per session; 25 sessions total.
  • Lectures based on notes (distributed in class and through the course webpage), with some additional materials, such as source papers or excerpts from textbooks
  • Student project talks scheduled during the last two weeks of the class

Out of Class

  • Problem sets
  • Course projects

Course Info

Instructor
As Taught In
Spring 2012
Level
Learning Resource Types
Lecture Notes
Projects with Examples
Instructor Insights