ML-MATLAB is a collection of MATLAB live scripts that illustrate concepts in machine learning. They can be used in the classroom or for self-study. These scripts have been used to teach the undergraduate course Computational Foundations of Machine Learning at Georgia Institute of Technology.

Each script usually ends with a section called "Taking it further" which offers some ideas for experimenting with the live script and learning more about the concepts that are illustrated. If you are a student, we encourage you to try out some of these ideas!

The goals behind these scripts are:

• Provide a way to understand statistics and machine learning concepts through seeing code (how equations translate into code) and also allow experimentation by modifying the data and algorithms.

• Show how easy it is to implement many machine learning algorithms directly in MATLAB, and provide a starting point for students' own implementations.

Since we primarily want to see the machine learning algorithms in their barest forms, the scripts generally do not use the machine learning tools built into MATLAB (an exception is the script for support vector machines). However, some examples of MATLAB tools may be included in the future for those who want to do more advanced prototyping of machine learning ideas in MATLAB.

Below, there are links to a web/html versions of the live scripts, for your convenience. If you want to experiment with the live scripts in MATLAB, download the scripts via github (in the live directory), or open the script directly in MATLAB Online using the buttons below.

The scripts are released with the MIT license. Feel free to modify and use the scripts as you need, while retaining the MIT license. Feedback and open-source contributions are welcome!

least_squares.html

estimate_variance.html

t_distribution.html

chi2_distribution.html

noise_variance_distribution.html

polynomial_regression.html

polyregr_bias_variance.html

polyregr_ridge.html

lasso_sparsity.html

lincombo_model.html

bayesian_inference.html

predictive_distribution.html

nonlinear_model.html

bayesian_simulation.html

mcmc.html

gaussian_processes.html

discriminants.html

logistic_regression.html

logregr_basisfun.html

support_vector_classifiers.html

support_vector_machines.html

neural_networks.html

stochastic_gradient_descent.html

nystrom.html