Supervised Learning — ZeroCourse

reading · we frame, you read MIT or the canonical taught · we author, no canonical fits ↺ spirals back to earlier lessons

Course locked

Complete Probability and Statistics first.

This course unlocks once you've finished its prerequisite. Open prerequisite →

Prerequisites

Linear Algebra not yet

Calculus and Optimization not yet

Probability and Statistics not yet

Complete prerequisites first to enroll.

reading · 35 min + ~249 min canonical

The Supervised Learning Pipeline: From Data to Deployed Model

Three full courses of ML math — linear algebra, calculus, probability — were the foundation. Now the machine learning itself begins. And it begins with the most useful, most-deplo…

primary Andrew Ng Machine Learning Specialization on Coursera

preview →

reading · 40 min + ~249 min canonical L1

Bias, Variance, Overfitting, and the Art of Generalization

The last lesson named the central drama of supervised learning — memorizing versus learning, fitting the training data versus generalizing to unseen data. This lesson is the theor…

primary Andrew Ng Machine Learning Specialization on Coursera

after L1 locked

reading · 35 min + ~249 min canonical L1

Linear and Logistic Regression: The Two Workhorses

Now the algorithms — and it's right to start here, with linear regression and logistic regression, because they are the two foundational models that everything else generalizes fr…

primary Andrew Ng Machine Learning Specialization on Coursera

after L2 locked

reading · 35 min + ~249 min canonical L1

Decision Trees and Ensembles: Why Boosting Wins on Tabular Data

Here is the most useful, least-hyped fact in applied machine learning: for the tabular data that most business problems involve, an ensemble of decision trees — gradient boosting,…

primary Andrew Ng Machine Learning Specialization on Coursera

after L3 locked

reading · 40 min + ~249 min canonical L1

The Rest of the Toolbox, and Honest Evaluation

Two things remain to complete your classical-ML foundation. First, the rest of the toolbox — three more algorithms, each embodying a distinct and instructive idea about how to lea…

primary Andrew Ng Machine Learning Specialization on Coursera

after L4 locked

exercise reading · 15 min L1

Exercises: Easy (Warm-up)

- [ ] Kaggle: Titanic — Survival prediction — Feature engineering (age bins, family size), logistic regression baseline - [ ] Implement linear regression from scratch in Python — …

after L5 locked

exercise reading · 30 min L1

Exercises: Medium (Core)

- [ ] Kaggle: House Prices — Advanced regression — Feature engineering, handle missing values, try Random Forest + XGBoost - [ ] Implement logistic regression from scratch — Sigmo…

after L6 locked

exercise reading · 45 min L1

Exercises: Hard (Stretch)

- [ ] Implement gradient boosting from scratch in Python — Sequential weak learners, residual fitting, learning rate - [ ] Implement SVM with SMO algorithm in Python — Quadratic p…

after L7 locked

exercise reading · 60 min L1

Mini-Project

- [ ] End-to-end ML pipeline — Pick a Kaggle dataset, full pipeline: EDA, feature engineering, model selection (compare 5+ models), hyperparameter tuning (Optuna), evaluation, wri…

after L8 locked

L10

project reading · 480 min L1

Project: ML Algorithm Implementations

Implement 3 algorithms from scratch in Python (no sklearn for the algorithm, only for data loading/evaluation): linear regression with gradient descent, logistic regression, and d…

after L9 locked

L11

project reading · 300 min L1

Project: Model Evaluation Dashboard

Build a Python script that takes any sklearn model and dataset, runs 5-fold cross-validation, computes metrics (accuracy, precision, recall, F1, AUC-ROC), plots learning curves an…

after L10 locked

L12

project reading · 600 min L1

Capstone: Kaggle Competition End-to-End

- [ ] Milestone 1: EDA — distributions, correlations, missing values, outliers (matplotlib/seaborn) - [ ] Milestone 2: Feature engineering — encoding, scaling, new features, clust…

after L11 locked

L13

assessment reading · 70 min L1

Self-Assessment: Supervised Learning

- [ ] Explain the bias-variance tradeoff. How does model complexity affect each? - [ ] What is regularization? Explain L1 (Lasso) vs L2 (Ridge) and when to use each. - [ ] Explain…

after L12 locked

13 lessons. Read in order; spiral back when you need to. By the end you'll have used the core ideas twice — once on the abstract, once on something you'll meet at work next week.