Supervised Machine Learning

Course 1 of 3 in the Machine Learning Specialization by Andrew Ng (DeepLearning.AI / Stanford).

Topics

Course 1 of 3 in the Machine Learning Specialization by Andrew Ng (DeepLearning.AI / Stanford).

Notes

 
Welcome to Machine Learning
machine-learning
supervised-learning

How artificial intelligence, machine learning, and deep learning fit together, why explicit rules run out, and what this first course covers.

Jul 3, 2026

Supervised Learning
machine-learning
supervised-learning

Supervised learning refers to algorithms that learn input-to-output mappings (x to y). The key characteristic is that you give the learning algorithm…

Jul 3, 2026

Unsupervised Learning
machine-learning
unsupervised-learning

In supervised learning, each example in the dataset is associated with an output label (for example, benign or malignant). In unsupervised learning,…

Jul 3, 2026

Linear Regression
machine-learning
supervised-learning
regression

Linear regression means fitting a straight line to your data. It is probably the most widely used learning algorithm in the world today. As you get…

Jul 3, 2026

Cost Function
machine-learning
supervised-learning
regression

How the cost function measures the gap between a linear model and the data, and why squared error is the standard choice for regression.

Jul 3, 2026

Gradient Descent
machine-learning
supervised-learning
regression

Gradient descent, the algorithm that finds the parameters at the bottom of the cost function, why it scales, and how the learning rate controls it.

Jul 5, 2026
 
Multiple Linear Regression
machine-learning
supervised-learning
regression

In univariate linear regression, you had a single feature (the size of a house) and predicted (the price). The model was:

Jul 6, 2026
 
Vectorization
machine-learning
supervised-learning
regression

In multiple linear regression, you learned to write the model compactly as:

Jul 6, 2026

Gradient Descent for Multiple Linear Regression
machine-learning
supervised-learning
regression

You have learned about gradient descent, multiple linear regression, and vectorization. This section ties those ideas together: gradient descent for a…

Jul 6, 2026

Feature Scaling
machine-learning
supervised-learning
regression

In the multiple variable lab, gradient descent ran slowly and predictions were still off after 1000 iterations. One reason is that the features had…

Jul 7, 2026

Checking Convergence and Choosing Learning Rate
machine-learning
supervised-learning
regression

When you run gradient descent, how can you tell if it is working? In other words, is it finding parameters and that are close to the minimum of the…

Jul 7, 2026

Feature Engineering and Polynomial Regression
machine-learning
supervised-learning
regression

The features you feed into a learning algorithm can have a huge impact on performance. For many real applications, choosing or engineering the right…

Jul 7, 2026

Classification
machine-learning
supervised-learning
classification

Last week you learned linear regression, which predicts a number. This week introduces classification, where the output can take on only one of a small…

Jul 7, 2026

Logistic Regression
machine-learning
supervised-learning
classification

For a long time, much of internet advertising was driven by a close cousin of logistic regression. Large websites used models like this to decide which…

Jul 7, 2026

Cost Function for Logistic Regression
machine-learning
supervised-learning
classification

The cost function tells you how well a set of parameters fits the training data. That score gives gradient descent a target to push toward better…

Jul 8, 2026

Gradient Descent for Logistic Regression
machine-learning
supervised-learning
classification

You now have the logistic regression model and the log loss cost function. The remaining step is to fit the parameters and by minimizing that cost. The…

Jul 8, 2026

The Problem of Overfitting
machine-learning
supervised-learning

You have now seen linear regression and logistic regression. Both work well on many tasks. Sometimes, though, a model can overfit the training data and…

Jul 8, 2026

Regularization
machine-learning
supervised-learning

The regularized cost function adds a lambda penalty on weights to shrink parameters, trade off fit vs simplicity, and reduce overfitting in linear models.

Jul 8, 2026

Regularized Linear and Logistic Regression
machine-learning
supervised-learning

Gradient descent with regularization for linear and logistic regression: the same lambda penalty on weights, weight decay, and unchanged bias updates.

Jul 9, 2026
No matching items
Back to top