Machine Learning for Beginners: From First Prediction to Learning Agents

Lesson 1: ML Lesson 0 | Learn Machine Learning: Your Course Roadmap

A short roadmap for the course. It outlines the path from a first prediction to learning agents and points to the full playlist and the separate four-hour version. The video ends with a prompt to try something and share your reasoning.

Play lesson

Lesson 2: ML Lesson 1 | Machine Learning Basics: Inputs, Targets & Success

Lesson 1 covers the basics of defining a useful prediction task. It shows how to identify inputs and targets and how to compare a model against a baseline. This sets up the rest of the course.

Play lesson

Lesson 3: ML Lesson 2 | Concept Learning: How Models Choose Between Rules

Lesson 2 follows a learner through a space of hypotheses. It explains why different rules can fit the same evidence and how a model chooses between them. This is concept learning in practice.

Play lesson

Lesson 4: ML Lesson 3 | Precision, Recall, Overfitting & Train/Test Splits

Lesson 3 focuses on evaluation. It walks through precision, recall, and the difference between types of mistakes. It also covers overfitting and how to design a fair train/test split.

Play lesson

Lesson 5: ML Lesson 4 | Decision Trees Explained: Predictions & Information Gain

Lesson 4 explains decision trees. It traces how a tree makes a prediction and calculates information gain to show why training prefers a useful split. The lesson stays with a small, traceable example.

Play lesson

Lesson 6: ML Lesson 5 | K-Nearest Neighbors: Distance, Voting & Feature Scaling

Lesson 5 covers k-nearest neighbors. It shows how to calculate a nearest-neighbor prediction and how feature scaling can change the answer. Distance and voting are the main ideas.

Play lesson

Lesson 7: ML Lesson 6 | Bayesian Learning: Base Rates, Bayes & Naive Bayes

Lesson 6 introduces Bayesian learning. It uses population counts to update beliefs and works through a simple naive Bayes classifier. Base rates and Bayes' rule are the focus.

Play lesson

Lesson 8: ML Lesson 7 | Neural Networks & Gradient Descent: A Worked Example

Lesson 7 is a worked example of a small neural network. It follows a weighted score, a gradient update, and the role of a hidden layer. Gradient descent is shown step by step.

Play lesson

Lesson 9: ML Lesson 8 | Learning Theory: Data, Model Complexity & Generalization

Lesson 8 covers learning theory. It works through a learning-theory guarantee while keeping its assumptions and limitations visible. Data, model complexity, and generalization are discussed together.

Play lesson

Lesson 10: ML Lesson 9 | Genetic Algorithms: Selection, Crossover & Mutation

Lesson 9 traces evolutionary search through selection, crossover, and mutation. It compares a promising candidate with an exact optimum. Genetic algorithms are the main topic.

Play lesson

Lesson 11: ML Lesson 10 | Rule Learning: Readable Predictions & Their Mistakes

Lesson 10 looks at rule learning. It follows readable prediction rules, compares candidates, and inspects errors and default behavior. The focus is on interpretable rules and where they go wrong.

Play lesson

Lesson 12: ML Lesson 11 | Explanation-Based Learning: Facts, Rules & Proofs

Lesson 11 covers explanation-based learning. It traces conclusions to their supporting facts and distinguishes missing evidence from negative evidence. Facts, rules, and proofs are the building blocks.

Play lesson

Lesson 13: ML Lesson 12 | Combining Prior Knowledge with Machine Learning

Lesson 12 shows how to combine prior knowledge with machine learning. It adds a knowledge-inspired feature, learns its weight, and compares the evidence with a plain model. The lesson keeps the comparison concrete.

Play lesson

Lesson 14: ML Lesson 13 | Reinforcement Learning & Your Machine Learning Project

Lesson 13 introduces reinforcement learning. It follows states, actions, rewards, and a value update, then ties the course together with a defensible model report. This is the final lesson in the series.

Play lesson