École d'hiver 2026 - Les Diablerets

1-4 février 2026

Eva Cantoni

Prof. Eva Cantoni, Université de Genève

Mme Caroline Gillardin, Coordinatrice CUSO

Prof. Francis Bach, INRIA, Paris, France. Subject : Learning Theory from First Principles

Prof. Tom Berrett, University of Warwick, UK. Subject : Non-parametric statistics

Prof. Philippe Naveau, LSCE, Gif sur Yvette Cedex, France. Subject : Statistics and Climate

Prof. Francis Bach, INRIA, Paris, France. Title : Learning Theory from First Principles

Abstract : Data have become ubiquitous in science, engineering, industry, and personal life, leading to the need for automated processing. Machine learning is concerned with making predictions from training examples and is used in all of these areas, in small and large problems, with a variety of learning models, ranging from simple linear models to deep neural networks. It has now become an important part of the algorithmic toolbox.

How can we make sense of these practical successes? Can we extract a few principles to understand current learning methods and guide the design of new techniques for new applications or to adapt to new computational environments? This is precisely the goal of learning theory and this series of two lectures, with a particular eye toward adaptivity to specific structures that make learning faster (such as smoothness of the prediction functions or dependence on low-dimensional subspaces).

Lectures : Consult the chapters 2-4-7-9 of the book "Learning Theory from First Principles" at the following address: Book. At the end of the course, the lecture notes (tablet) will be made available to the participants.

Notes

Board1

Board2

Board3

Prof. Tom Berrett, University of Warwick, UK.Title: Hypothesis testing: Randomisation, privacy and minimax optimality

Abstract: In these lectures I will discuss recent advances in the methodology and theory of hypothesis testing, with a focus on finite-sample guarantees, robustness and reliability. Widely deployed across scientific research, hypothesis testing is a cornerstone of statistical thinking that allows us to make interpretable statements of uncertainty quantification. It also underpins much of modern statistical optimality theory. In the first lecture we will consider problems with simple null hypotheses, building up from the classical Neyman--Pearson lemma for two-point testing to surprising new results in instance optimality for goodness-of-fit testing. In the second lecture we will consider new developments in the application and theory of permutation testing, seeing how this classical technique can provide both uniform type I error control and optimal power for certain problems with composite null hypotheses. Finally, in the third lecture, we will focus on issues of robustness and data security as we survey hypothesis testing problems in the field of local differential privacy, seeing how we must adapt traditional approaches to protect data subjects' personal information.

Lectures

Prof. Philippe Naveau, LSCE, Gif sur Yvette Cedex, France. Title : Statistics and Climate

Abstract : This lecture will focus on the statistical analysis of extreme events in a changing climate and their impacts on related risks (floods, droughts, etc). Classical univariate, multivariate and spatial extreme value theory models will be explained in order to assess climate risk in different setups. Different key indices to capture dependencies among extremes will be studied. If time allowed, various inference techniques (frequencies, Bayesian, neural network) will be covered via different climatological hands-on examples. 

Lecture1

Lecture2

Lecture3

Le code R est dispo sur : https://sharebox.lsce.ipsl.fr/index.php/s/osljdVH063cRwok

Les articles liés à ces trois présentations se trouvent sur  : https://sharebox.lsce.ipsl.fr/index.php/s/07bosRr1dxgXss5

Breakfast, lunch and dinner : Hotel Glacier 3000

Monday night, "raclette evening" in Chalet Téléthon (5 minutes walking from The Glacier Hotel 3000) at 7:30 PM

This program may be changed

<figure class="table">

<figure class="table">

<figure class="table">

<figure class="table">

<figure class="table">

</figure>

</figure>

</figure>

</figure>

</figure>

</figure>

Hôtel Glacier 3000, Les Diablerets

<figure class="table">

<figure class="table">

<figure class="table">

<figure class="table">

<figure class="table">

</figure>

</figure>

</figure>

</figure>

</figure>

</figure>

Versement sur compte postal (payment into postal account) :

CUSO

CCP 12-1873-8

Neuchâtel

BIC : POFICHBEXXX

IBAN : CH0509000000120018738. Merci d'écrire votre nom suivi du no" 26220002 " lors du paiement. Thank you to write your name on the payment wording and nb" 26220002".

IMPORTANT : si vous vous inscrivez, et que vous avez un empêchement, veuillez nous contacter dès que possible. Sinon le prix de toutes les nuitées à l'hôtel vous seront intégralement facturées si l'hôtel le demande. If you register and then have to cancel, please contact us as soon as possible. Otherwise, the full cost of all hotel nights will be charged to you if requested by the hotel.

Tarif :

Non CUSO (ETHZ, etc)

Non CUSO (ETH) double room: 1300 CHF (price may fluctuate depending on the nb of registrations)

Non CUSO (ETH) single room: 1400 CHF (price may fluctuate depending on the nb of registrations)

You are responsible for your own transport costs

CUSO (UNINE, UNIGE, UNIL, UNIFR, UNIBE, IHEID)

Doctorant-e CUSO chambre double: 200 CHF

Doctorant-e CUSO chambre simple: 350 CHF

Post-doctorant-e CUSO chambre double: 300 CHF

Post-doctorant-e CUSO chambre simple: 450 CHF

Professeur-e CUSO chambre double: 400 CHF

Professeur-e CUSO chambre simple: 550 CHF

Non CUSO privé-e chambre double: 1300 CHF

Non CUSO privé-e chambre simple: 1400 CHF

Lors de votre inscription, merci de bien vouloir indiquer dans la zone commentaire si vous désirez une chambre simple, ou double et le nom de la personne avec qui vous souhaiteriez partager votre chambre. Dans le cas où rien n'est indiqué, une chambre simple sera réservée.

When registering, please indicate in the comment section whether you would like a single or double room, and the name of the person you wish to share your room with. If no preference is specified, a single room will be reserved.

36