CS 189 at UC Berkeley

# Introduction to Machine Learning

Lectures: M-Th 6:30-8:00 PM in VLSB 2060

## Instructor Marc Khoury

khoury [at] eecs.berkeley.edu

## Instructor Brijen Thananjeyan

bthananjeyan [at] berkeley.edu

Office Hours: Wed 11AM-12PM (Soda 341B)

### Week 1 Overview

## Introduction, Linear Classifiers, Perceptron Learning, and the support vector classifier

### Week 2 Overview

## Machine learning abstractions, Decision theory, and Gaussian discriminant analysis

### Week 3 Overview

## Eigenvectors, Quadratic Forms, Normal distributions, Regression, Newton's method and logistic regression

### Week 4 Overview

## Statistical justifications for regression, Ridge regression

## Notes

There is no textbook for this class, and we do not plan on posting lecture slides. Instead, see Shewchuk's lecture notes and "A Comprehensive Guide to Machine Learning" on our Resources page, which cover a large portion of the material covered in lecture. We will follow Shewchuk's lecture notes closely. Notes are not a substitute for going to lecture, as additional material may be covered in lecture.

## Discussions

The discussion sections may cover new material and will give you additional practice solving problems. See Syllabus for more information.

- Discussion 1: Support Vector Machines (solution)
- Discussion 2: Review (solution)
- Discussion 3: MLE, MAP, Bayesian Decision Theory (solution)
- Discussion 4: Gaussian Discriminant Analysis, MLE (solution)
- Discussion 5: Gaussian Isocontours, LDA (solution)
- Discussion 6: Regression, Pseudoinverse, SGD (solution)
- Discussion 7: Bias and Variance (solution)

Expand