This repository contains slides and exercises for Epita students of the FTML and PTML courses.
This folder contains the slides of the courses and some related exercises.
- Introduction
- Mathematical tools for ML, Supervised learning,
- Risk decomposition, Optimization in ML, Ordinary least squares.
- Ridge regression, Classification, Logistic regression
- Unsupervised learning: clustering, dimensionality reduction (PCA)
- Gradient algorithms, probabilistic modelling
- Support vector machines, Kernel methods
- Decision tree learning, Ensemble learning
- Scoring and cross-validation, neural networks I
- Neural networks II, statistical learning
- Local methods, Sparse methods and variable selection, adaptivity
- Bayesian learning, latent variables
This folder contains the practical sessions.
- Reminders/Introduction to Python 3, polynomial regression, Ordinary least squares
- Ridge regression, logistic regression, cross validation I
- Gradient descent (GD) and extensions (on OLS)
- Stochastic gradient descent (SGD) and extensions (on logistic regression)
- Stochastic average gradient (SAG) (on logistic regression)
- Bayesian learning
- Model selection, PCA, kernel PCA
- Neural networks
- Nonlinear dimensionality reduction
- Modern topics on gradient descent