ENBIS-18 Post-Conference Event: Joint ECAS-ENBIS 1-Day Summer School

6 September 2018; 08:30 – 17:00; IECL, Faculté des Sciences, Vandoeuvre-lès-Nancy

This 1-day course is a joint initiative from ENBIS and ECAS (ecas.fenstats.eu), which provides courses since 1987 to achieve training in special areas of statistics for both researchers and teachers for universities and professionals in industry field.

The 1-day course is done under the umbrella of the ENBIS-18 conference in Nancy. You can find more information about the conference here.

 

Introduction to Deep Learning

This course is an introduction to deep learning. We will present deep learning from a historical perspective ranging from the formal neuron up to recent models from both supervised and unsupervised learning. Special emphasis will be on applications domains (image, audio, NLP, games), architectures design, generative learning and non-convex optimization.

LECTURER

Stéphane Canu is a Professor of the LITIS research laboratory and of the information technology department, at the National institute of applied science in Rouen (INSA).
 He has been the dean of the computer engineering department he create in 1998 until 2002 when he was named director of the computing service and facilities unit. In 2004 he join for one sabbatical year the machine learning group at ANU/NICTA (Canberra) with Alex Smola and Bob Williamson.  In the last five years, he has published approximately thirty papers in refereed conference proceedings or journals in the areas of theory, algorithms and applications using kernel machines learning algorithm and other flexible regression methods.  His research interests include deep learning, kernels machines, regularization, machine learning applied to signal processing, pattern classification, factorization for recommander systems and learning for context aware applications.

 

An Introduction to Generalized Additive Models

Generalized Additive Models (GAMs) models are an extension of traditional parametric regression models, which have proved highly useful for both predictive and inferential purposes in a wide variety of scientific and commercial applications. One reason behind the popularity of GAMs is that they strike an interesting balance between flexibility and interpretability, while being able to handle large data sets. The mgcv R package is arguably the state-of-the-art tool for fitting such models, hence the first half of this tutorial will introduce GAMs and mgcv, in the context of electricity demand forecasting. The second part of the tutorial will show how traditional GAMs can be extended to quantile GAMs, and how the latter can be fitted using the qgam R package. By the end of the tutorial the attendees should be able to build, fit and visualize traditional or quantile GAM models, using a combination of the mgcv, qgam and mgcViz R packages. This tutorial is aimed at a broad audience of statistical modellers, interested in using GAMs for predictive or inferential purposes. The models which will be presented in the tutorial have a very wide range of applicability, hence they should be of interest to practitioners in business intelligence, ecology, linguistics, epidemiology and geoscience to name a few.

LECTURER

Matteo Fasiolo works as a research associate at the University of Bristol. His current research is concerned with extending Generalized Additive Models (GAMs), with particular focus on electricity load forecasting applications. He is the authors of the `qgam` R package, which provides fitting methods for quantile regression GAMs, and he is currently developing, in collaboration with Raphaël Nedellec at EDF, the `mgcViz` R package, which offers new visualization tools for GAMs. He is also the author of the `mvnfast`, `synlik` and `esaddle` R packages on CRAN. Matteo has an MEng in Industrial Engineering from the University of Udine (Italy), an MSc in Financial Engineering from the University of London (Birkbeck College) and a PhD in Statistics from the University of Bath, where he was supervised by Prof. Simon N. Wood.