Machine Learning, from Theory to Practice

Course @ Ecole polytechnique by Erwan Le Pennec and Francoise d'Alché-Buc

Course 1

Large review of machine learning principles

Course 2

SVM - SVR, theory of RKHS, supervised and unsupervised learning in a RKHS

Course 3

Unsupervised learning, dimension reduction

Course 4

Tree & ensemble methods: decision/regression trees, ensemble (bagging, random forest, adaboost, gradient boosting, anyboost)

Course 5

Graphs in ML I: spectral clustering, transductive learning, semi-supervised learning

Course 6

Graphs in ML II: KNN for collaborative filtering, matrix factorization (with link with PCA, convex formulation & proximal gradient descent), applications to image completion

Course 7

Kernels and margin bounds: review on kernels, margin theory

Course 8

Introduction to Neural Networks: perceptron, multlayer perceptron, backpropagation algorithm

Course 9

Feature design (renormalization, dictionnary learning, feature encoding, quantization/binarization, hashing, pooling) and text/image mining (texts and bag of words, word vectors, convolutional networks, recurrent neural networks for text)