Machine Learning

Machine learning notes organized by topic, covering the Machine Learning Specialization and further modules such as federated learning.

These notes are organized by topic. Each module cites its own reference material, so some modules follow a Coursera course while others draw on short courses or other sources.

Publication Dates and Currency

Course material ages, so every module below records when it was first published. The Machine Learning Specialization launched in June 2022 as the rebuilt successor to Andrew Ng’s 2012 Stanford course, trading Octave for Python, NumPy, scikit-learn, and TensorFlow. All three courses shipped together, and these notes follow that build. Read library APIs and any “current practice” framing as a snapshot of the publication year rather than of today. Lab code is the exception, ported to the current Python stack, with a callout on each lab page naming what changed.

Specialization Roadmap

The first three modules cover the Machine Learning Specialization by Andrew Ng (DeepLearning.AI / Stanford).

Course 1: Supervised Machine Learning

Foundations of machine learning, including linear regression, logistic regression (classification), gradient descent, and regularization.

Reference: https://www.coursera.org/learn/machine-learning

Published: June 2022. This Coursera address previously hosted Andrew Ng’s original 2012 machine learning course.

Course 2: Advanced Learning Algorithms

More powerful models and practical techniques, including neural networks, decision trees, and advice for applying machine learning (bias and variance, debugging).

Reference: https://www.coursera.org/learn/advanced-learning-algorithms

Published: June 2022.

Course 3: Unsupervised Learning, Recommenders, Reinforcement Learning

Important algorithms beyond supervised learning, including clustering (K-means), anomaly detection, recommender systems, and reinforcement learning.

Reference: https://www.coursera.org/learn/unsupervised-learning-recommenders-reinforcement-learning

Published: June 2022.

Beyond the Specialization

Federated Learning

Training models on distributed data without centralizing it, including the Flower framework, tuning federated systems, differential privacy, and bandwidth.

Reference: https://www.deeplearning.ai/short-courses/intro-to-federated-learning/

Published: July 2024.