Statistical Learning II
Welcome to the course page for Statistical Learning II. Here you’ll find all the course information and materials.
Course Information
- Instructors:
- Dr. Bruno Loureiro (main instructor)
- Leonardo Defilippis (TA)
- Email: bruno.loureiro@di.ens.fr
- Semester: Fall 2024
- Class Times: Wednesdays, 08:30 AM - 11:45 AM
- Location: Check at the Dauphine ENT.
Course Description
This course is a continuation of the “Statistical Learning I” course taught at the 2nd year (L2) of the Double Licence Intelligence Artificielle et Sciences des Organisations (IASO) undergraduate from Université Paris-Dauphine.
Our goal is to build a basic understanding of the mathematics behind some of the classical machine learning algorithms, such as:
- Least squares regression
- Ridge regression
- LASSO
- Logistic regression
- Kernel methods
- Random features regression
- PCA
As well as the most widely used optimisation algorithms used to train them, such as stochastic gradient descent.
Recommended literature
The material in this course takes inspiration from the following excellent ressources:
- Bach, Francis. Learning theory from first principles. MIT press, 2024.
- Hastie, Trevor, et al. The elements of statistical learning: data mining, inference, and prediction. Vol. 2. New York: springer, 2009.
- Murphy, Kevin P. Machine learning: a probabilistic perspective. MIT press, 2012.
- Wasserman, Larry. All of statistics: a concise course in statistical inference. Springer Science & Business Media, 2013.
Course Schedule
Date | Lecture Topic | Materials |
---|---|---|
Sept 11 | - Introduction - Recap of Linear Algebra | Slides Jupyter Notebook |
Sept 18 | - Recap of Probability - Supervised Learning | Slides Jupyter Notebook |
Sept 25 | - Supervised Learning (continued) | Slides |
Oct 02 | - Least-squares regression | Slides |
Oct 09 | - Least-squares regression (continued) | Slides |
Oct 16 | - Bias-variance decomposition | Slides |
Oct 23 | Midterm exam | |
Oct 30 | Reading week (no class) | |
Nov 06 | - Ridge regression | Slides |
Nov 13 | - Ridge regression (continued) | Slides |