Introduction to machine learning

Material

Outline

I. Supervised learning

  1. Discriminant analysis, logistic regression and boosting
    Practical 1.

  2. Support vector classification and regression
    Movie: polynomial kernel in action.
    Practical 2.
    SVM GUI.

  3. Nonparametric and ensemble methods
    Practical 3.

II. Clustering

  1. Gaussian mixtures and k-means
    Practical 4.

  2. Spectral clustering and agglomerative approaches
    Practical 5.
    Movie: DBSCAN in action.

  3. Clustering evaluation
    Practical 6.

III. Dimensionality reduction

  1. Principal component analysis
    Practical 7.

  2. Random projection and nonlinear methods

Data challenge

You should join the Paris Fire Brigade data challenge (the password is 4586), which ends on Tuesday December 31, 2019, and to fill out this form.

Part of your final mark will be:

  1. your ranking (at least one submission is mandatory);

  2. your 2-page report describing:

  • the data preprocessing;

  • the model selection phase;

  • your final model;

  • the impediments you faced and your achievements.

Your report has to be uploaded on this remote repository by Friday January 10, 2019.

Exam