Machine Learning

Course sites

• For technical questions: Please read/post publicly on Piazza
• For personal or administrative questions: Please post privately on Piazza
• For grades on submitted work: Please check Canvas

Course Overview

• Probabilistic foundations of machine learning; offered in fall semesters; 12 units (3-0-9)
• Prerequisites:
  • Introductory machine learning (at the level of 6.390 or 6.370 or 6.C01)
  • Linear algebra (at the level of 18.06 or 18.C06)
  • Probability (at the level of 6.3700, 6.3800, or 18.600)
• Brief description: Probabilistic thinking is critical to understanding machine learning, in techniques ranging from classic linear models to modern deep networks. We will study model representation, generalization, learning algorithms, and model-selection with mathematical rigor as well as an emphasis on how to apply these methods in applications with real-world consequence.
• A syllabus can be found here

Course Components

We will have weekly lectures and problem sessions (schedule details can be found here); complementing those are written homework, small projects, a midterm and a final exam.

Course Number Change

Since Fall 22, all MIT EECS subjects have been renumbered (rationale and details can be found here). This subject used to be called 6.867; moving forward, it will be 6.790, eventually. But for registration purposes, please register for 6.7900 (note the extra trailing zero) during this current transition phase.