Machine-Learning-Tutorial

given during ADA IX Summer School held on 20-22 May 2018 in Valencia, Spain

Authors

Emille E. O. Ishida - CNRS/LPC-Clermont, France
email, twitter, website

Alexandre Boucaud - Paris-Saclay Center for Data Science, France
email, twitter, github

Introduction

Session 1 - 20 May 2018, 11:15h - 13:00h

Basic principles of Machine Learning
Supervised and unsupervised learning

slides

Regression example: Boston dataset
Regression example: Photometric redshift estimation

Tutorial session: Machine Learning in practice

Regression - notebook

Classification - notebook

From NN to CNN

Session 2 - 20 May 2018, 14:00h - 15:45h

Hands on deep learning

Neurons and backpropagation
Convolutional neural networks
In practice
Common optimizations

slides* - references - solution of exercise

*use arrow keys to navigate between slides

Cooking a simple neural network library

notebook - solutions

Beyond text-book Machine Learning

Session 3 - 20 May 2018, 17:15h - 18:15h

Adaptive
Reinforcement
Self-trained

slides

Extra Material

Supernova Photometric Classification as a Data Challenge

RAMP starting-kit
PLAsTiCC - Photometric LSST Astronomical Time-series Classification Challenge
References