From Data to Models

Introduction to Machine Learning

The foundations of machine learning — regression, classification, neural networks, and unsupervised learning — from first examples to working models.

Lectures

8 modules

These lectures are incomplete and are being added to as the semester goes on. What is here will keep changing, and slides may appear, move or be rewritten between one week and the next.

Introduction and Logistics

Why build intelligent machines, a brief history of AI, what this course is about, and how it works as an experience.

Learning, Examples, and Data

An overview of the various forms of data and examples of their use in machine learning.

PS0

Problem Set 0

Getting set up on the site, and the three conversations that make up Problem Set 0.

Math and Python: Your Computational Toolkit

Vectors, matrices, gradients and pandas, with every idea written twice: once in notation, once in code.

Feature Engineering and Generalization

Feature maps, the bias-variance trade-off, train/validation/test, cross validation, and regularization.

Classification & Logistic Regression (snapshot)

Keynote snapshot of the original slides, kept for reference beside the native rebuild.

MLE and Generalized Linear Models

Least squares and cross-entropy as one idea: maximum likelihood, the exponential family, and the update rule they share.

Generative Learning and Language

Words as vectors, bag of words, naive Bayes by counting, Bayes rule, and the line from n-grams to language models.