2021
Activity report
Project-Team
STATIFY
RNSR: 202023582A
Research center
In partnership with:
CNRS, Institut polytechnique de Grenoble
Team name:
Bayesian and extreme value statistical models for structured and high dimensional data
In collaboration with:
Laboratoire Jean Kuntzmann (LJK)
Domain
Applied Mathematics, Computation and Simulation
Theme
Optimization, machine learning and statistical methods
Creation of the Project-Team: 2020 April 01

# Keywords

• A3.1.1. Modeling, representation
• A3.1.4. Uncertain data
• A3.3.2. Data mining
• A3.3.3. Big data analysis
• A3.4.1. Supervised learning
• A3.4.2. Unsupervised learning
• A3.4.4. Optimization and learning
• A3.4.5. Bayesian methods
• A3.4.7. Kernel methods
• A5.3.3. Pattern recognition
• A5.9.2. Estimation, modeling
• A6.2. Scientific computing, Numerical Analysis & Optimization
• A6.2.3. Probabilistic methods
• A6.2.4. Statistical methods
• A6.3. Computation-data interaction
• A6.3.1. Inverse problems
• A6.3.3. Data processing
• A6.3.5. Uncertainty Quantification
• A9.2. Machine learning
• A9.3. Signal analysis
• B1.2.1. Understanding and simulation of the brain and the nervous system
• B2.6.1. Brain imaging
• B3.3. Geosciences
• B3.4.1. Natural risks
• B3.4.2. Industrial risks and waste
• B3.5. Agronomy
• B5.1. Factory of the future
• B9.5.6. Data science
• B9.11.1. Environmental risks

# 1 Team members, visitors, external collaborators

## Research Scientists

• Florence Forbes [Team leader, Inria, Senior Researcher, HDR]
• Sophie Achard [CNRS, Senior Researcher, HDR]
• Julyan Arbel [Inria, Researcher, HDR]
• Pedro Luiz Coelho Rodrigues [Inria, from Oct 2021, Starting Faculty Position]
• Stephane Girard [Inria, Senior Researcher, HDR]
• Pierre Wolinski [Univ Grenoble Alpes, Starting Research Position, from Oct 2021]

## Faculty Members

• Jean-Baptiste Durand [Institut polytechnique de Grenoble, Associate Professor]
• Jonathan El-Methni [Université de Paris, Associate Professor]
• Olivier Francois [Institut polytechnique de Grenoble, Professor, from Sep 2021]

## Post-Doctoral Fellow

• Pascal Dkengne Sielenou [Inria, until Feb 2021]

## PhD Students

• Louise Alamichel [Univ Grenoble Alpes, from Oct 2021]
• Yuchen Bai [Univ Grenoble Alpes, from Oct 2021]
• Meryem Bousebata [Univ Grenoble Alpes]
• Daria Bystrova [Univ Grenoble Alpes]
• Lucrezia Carboni [Univ Grenoble Alpes]
• Alexandre Constantin [Univ Grenoble Alpes]
• Benoit Kugler [Univ Grenoble Alpes]
• Hana Lbath [Univ Grenoble Alpes]
• Julia Linhart [Université Paris-Saclay, Nov 2021, Inria Saclay]
• Minh Tri [Invensens]
• Theo Moins [Inria]
• Giovanni Poggiato [Univ Grenoble Alpes]

## Technical Staff

• Pascal Dkengne Sielenou [Inria, Engineer, from Mar 2021]
• Antoine Lesieur [Inria, Engineer, from Apr 2021 until Sep 2021]

## Interns and Apprentices

• Louise Alamichel [Univ Grenoble Alpes, from Apr 2021 until Jul 2021]
• Mahtab Khademalhosseini [Inria, from Feb 2021 until Jul 2021]
• Khalil Leachouri [Inria, from Jun 2021 until Aug 2021]
• Hichem Saghrouni [Ecole normale supérieure Paris-Saclay, from Aug 2021 until Sep 2021]

• Geraldine Christin [Inria, until Aug 2021]

## Visiting Scientist

• Trung Tin Nguyen [Univ de Caen Basse-Normandie, Jan 2021]

# 2 Overall objectives

The statify team focuses on statistics. Statistics can be defined as a science of variation where the main question is how to acquire knowledge in the face of variation. In the past, statistics were seen as an opportunity to play in various backyards. Today, the statistician sees his own backyard invaded by data scientists, machine learners and other computer scientists of all kinds. Everyone wants to do data analysis and some (but not all) do it very well. Generally, data analysis algorithms and associated network architectures are empirically validated using domain-specific datasets and data challenges. While winning such challenges is certainly rewarding, statistical validation lies on more fundamentally grounded bases and raises interesting theoretical, algorithmic and practical insights. Statistical questions can be converted to probability questions by the use of probability models. Once certain assumptions about the mechanisms generating the data are made, statistical questions can be answered using probability theory. However, the proper formulation and checking of these probability models is just as important, or even more important, than the subsequent analysis of the problem using these models. The first question is then how to formulate and evaluate probabilistic models for the problem at hand. The second question is how to obtain answers after a certain model has been assumed. This latter task can be more a matter of applied probability theory, and in practice, contains optimization and numerical analysis.

The statify team aims at bringing strengths, at a time when the number of solicitations received by statisticians increases considerably because of the successive waves of big data, data science and deep learning. The difficulty is to back up our approaches with reliable mathematics while what we have is often only empirical observations that we are not able to explain. Guiding data analysis with statistical justification is a challenge in itself. statify has the ambition to play a role in this task and to provide answers to questions about the appropriate usage of statistics.

Often statistical assumptions do not hold. Under what conditions then can we use statistical methods to obtain reliable knowledge? These conditions are rarely the natural state of complex systems. The central motivation of statify is to establish the conditions under which statistical assumptions and associated inference procedures approximately hold and become reliable.

However, as George Box said "Statisticians and artists both suffer from being too easily in love with their models". To moderate this risk, we choose to develop, in the team, expertise from different statistical domains to offer different solutions to attack a variety of problems. This is possible because these domains share the same mathematical food chain, from probability and measure theory to statistical modeling, inference and data analysis.

Our goal is to exploit methodological resources from statistics and machine learning to develop models that handle variability and that scale to high dimensional data while maintaining our ability to assess their correctness, typically the uncertainty associated with the provided solutions. To reach this goal, the team offers a unique range of expertise in statistics, combining probabilistic graphical models and mixture models to analyze structured data, Bayesian analysis to model knowledge and regularize ill-posed problems, non-parametric statistics, risk modeling and extreme value theory to face the lack, or impossibility, of precise modeling information and data. In the team, this expertise is organized to target five key challenges:

• 1.
Models for high dimensional, multimodal, heterogeneous data;
• 2.
Spatial (structured) data science;
• 3.
Scalable Bayesian models and procedures;
• 4.
Understanding mathematical properties of statistical and machine learning methods;
• 5.
The big problem of small data.

The first two challenges address sources of complexity coming from data, namely, the fact that observations can be: 1) high dimensional, collected from multiple sensors in varying conditions i.e. multimodal and heterogeneous and 2) inter-dependent with a known structure between variables or with unknown interactions to be discovered. The other three challenges focus on providing reliable and interpretable models: 3) making the Bayesian approach scalable to handle large and complex data; 4) quantifying the information processing properties of machine learning methods and 5) allowing to draw reliable conclusions from datasets that are too small or not large enough to be used for training machine/deep learning methods.

These challenges rely on our four research axes:

• 1.
Models for graphs and networks;
• 2.
Dimension reduction and latent variable modeling;
• 3.
Bayesian modeling;
• 4.
Modeling and quantifying extreme risk.

In terms of applied work, we will target high-impact applications in neuroimaging, environmental and earth sciences.

# 3 Research program

## 3.1 Mixture models

Participants: Jean-Baptiste Durand, Florence Forbes, Stephane Girard, Julyan Arbel, Olivier Francois, Daria Bystrova, Giovanni Poggiato, Benoit Kugler, Alexandre Constantin, Louise Alamichel.

Keywords: Key-words: mixture of distributions, EM algorithm, missing data, conditional independence, statistical pattern recognition, clustering, unsupervised and partially supervised learning..

In a first approach, we consider statistical parametric models, $\theta$ being the parameter, possibly multi-dimensional, usually unknown and to be estimated. We consider cases where the data naturally divides into observed data $y=\left\{{y}_{1},...,{y}_{n}\right\}$ and unobserved or missing data $z=\left\{{z}_{1},...,{z}_{n}\right\}$. The missing data ${z}_{i}$ represents for instance the memberships of one of a set of $K$ alternative categories. The distribution of an observed ${y}_{i}$ can be written as a finite mixture of distributions,

$\begin{array}{c}\hfill f\left({y}_{i};\theta \right)=\sum _{k=1}^{K}P\left({z}_{i}=k;\theta \right)f\left({y}_{i}\mid {z}_{i};\theta \right)\phantom{\rule{0.277778em}{0ex}}.\end{array}$ 1

These models are interesting in that they may point out hidden variables responsible for most of the observed variability and so that the observed variables are conditionally independent. Their estimation is often difficult due to the missing data. The Expectation-Maximization (EM) algorithm is a general and now standard approach to maximization of the likelihood in missing data problems. It provides parameter estimation but also values for missing data.

Mixture models correspond to independent ${z}_{i}$'s. They have been increasingly used in statistical pattern recognition. They enable a formal (model-based) approach to (unsupervised) clustering.

## 3.2 Graphical and Markov models

Participants: Jean-Baptiste Durand, Florence Forbes, Julyan Arbel, Sophie Achard, Olivier Francois, Mariia Vladimirova, Lucrezia Carboni, Hana Lbath, Minh-tri Le, Yuchen Bai.

Keywords: Key-words: graphical models, Markov properties, hidden Markov models, clustering, missing data, mixture of distributions, EM algorithm, image analysis, Bayesian inference..

Graphical modelling provides a diagrammatic representation of the dependency structure of a joint probability distribution, in the form of a network or graph depicting the local relations among variables. The graph can have directed or undirected links or edges between the nodes, which represent the individual variables. Associated with the graph are various Markov properties that specify how the graph encodes conditional independence assumptions.

It is the conditional independence assumptions that give graphical models their fundamental modular structure, enabling computation of globally interesting quantities from local specifications. In this way graphical models form an essential basis for our methodologies based on structures.

The graphs can be either directed, e.g. Bayesian Networks, or undirected, e.g. Markov Random Fields. The specificity of Markovian models is that the dependencies between the nodes are limited to the nearest neighbor nodes. The neighborhood definition can vary and be adapted to the problem of interest. When parts of the variables (nodes) are not observed or missing, we refer to these models as Hidden Markov Models (HMM). Hidden Markov chains or hidden Markov fields correspond to cases where the ${z}_{i}$'s in (1) are distributed according to a Markov chain or a Markov field. They are a natural extension of mixture models. They are widely used in signal processing (speech recognition, genome sequence analysis) and in image processing (remote sensing, MRI, etc.). Such models are very flexible in practice and can naturally account for the phenomena to be studied.

Hidden Markov models are very useful in modelling spatial dependencies but these dependencies and the possible existence of hidden variables are also responsible for a typically large amount of computation. It follows that the statistical analysis may not be straightforward. Typical issues are related to the neighborhood structure to be chosen when not dictated by the context and the possible high dimensionality of the observations. This also requires a good understanding of the role of each parameter and methods to tune them depending on the goal in mind. Regarding estimation algorithms, they correspond to an energy minimization problem which is NP-hard and usually performed through approximation. We focus on a certain type of methods based on variational approximations and propose effective algorithms which show good performance in practice and for which we also study theoretical properties. We also propose some tools for model selection. Eventually we investigate ways to extend the standard Hidden Markov Field model to increase its modelling power.

## 3.3 Functional Inference, semi- and non-parametric methods

Participants: Julyan Arbel, Daria Bystrova, Giovanni Poggiato, Stephane Girard, Florence Forbes, Pedro Coelho Rodrigues, Pascal Dkengne Sielenou, Meryem Bousebata, Theo Moins, Pierre Wolinski, Sophie Achard.

Keywords: Key-words: dimension reduction, extreme value analysis, functional estimation..

We also consider methods which do not assume a parametric model. The approaches are non-parametric in the sense that they do not require the assumption of a prior model on the unknown quantities. This property is important since, for image applications for instance, it is very difficult to introduce sufficiently general parametric models because of the wide variety of image contents. Projection methods are then a way to decompose the unknown quantity on a set of functions (e.g. wavelets). Kernel methods which rely on smoothing the data using a set of kernels (usually probability distributions) are other examples. Relationships exist between these methods and learning techniques using Support Vector Machine (SVM) as this appears in the context of level-sets estimation (see section 3.3.2). Such non-parametric methods have become the cornerstone when dealing with functional data 92. This is the case, for instance, when observations are curves. They enable us to model the data without a discretization step. More generally, these techniques are of great use for dimension reduction purposes (section 3.3.3). They enable reduction of the dimension of the functional or multivariate data without assumptions on the observations distribution. Semi-parametric methods refer to methods that include both parametric and non-parametric aspects. Examples include the Sliced Inverse Regression (SIR) method 94 which combines non-parametric regression techniques with parametric dimension reduction aspects. This is also the case in extreme value analysis91, which is based on the modelling of distribution tails (see section 3.3.1). It differs from traditional statistics which focuses on the central part of distributions, i.e. on the most probable events. Extreme value theory shows that distribution tails can be modelled by both a functional part and a real parameter, the extreme value index.

### 3.3.1 Modelling extremal events

Extreme value theory is a branch of statistics dealing with the extreme deviations from the bulk of probability distributions. More specifically, it focuses on the limiting distributions for the minimum or the maximum of a large collection of random observations from the same arbitrary distribution. Let ${X}_{1,n}\le ...\le {X}_{n,n}$ denote $n$ ordered observations from a random variable $X$ representing some quantity of interest. A ${p}_{n}$-quantile of $X$ is the value ${x}_{{p}_{n}}$ such that the probability that $X$ is greater than ${x}_{{p}_{n}}$ is ${p}_{n}$, i.e.$P\left(X>{x}_{{p}_{n}}\right)={p}_{n}$. When ${p}_{n}<1/n$, such a quantile is said to be extreme since it is usually greater than the maximum observation ${X}_{n,n}$.

To estimate such quantiles therefore requires dedicated methods to extrapolate information beyond the observed values of $X$. Those methods are based on Extreme value theory. This kind of issue appeared in hydrology. One objective was to assess risk for highly unusual events, such as 100-year floods, starting from flows measured over 50 years. To this end, semi-parametric models of the tail are considered:

$P\left(X>x\right)={x}^{-1/\theta }\ell \left(x\right),\phantom{\rule{0.277778em}{0ex}}x>{x}_{0}>0,$ 2

where both the extreme-value index $\theta >0$ and the function $\ell \left(x\right)$ are unknown. The function $\ell$ is a slowly varying function i.e. such that

$\frac{\ell \left(tx\right)}{\ell \left(x\right)}\to 1\phantom{\rule{4.pt}{0ex}}\text{as}\phantom{\rule{4.pt}{0ex}}x\to \infty$ 3

for all $t>0$. The function $\ell \left(x\right)$ acts as a nuisance parameter which yields a bias in the classical extreme-value estimators developed so far. Such models are often referred to as heavy-tail models since the probability of extreme events decreases at a polynomial rate to zero. It may be necessary to refine the model (2,3) by specifying a precise rate of convergence in (3). To this end, a second order condition is introduced involving an additional parameter $\rho \le 0$. The larger $\rho$ is, the slower the convergence in (3) and the more difficult the estimation of extreme quantiles.

More generally, the problems that we address are part of the risk management theory. For instance, in reliability, the distributions of interest are included in a semi-parametric family whose tails are decreasing exponentially fast. These so-called Weibull-tail distributions 10 are defined by their survival distribution function:

$P\left(X>x\right)=exp\left\{-{x}^{\theta }\ell \left(x\right)\right\},\phantom{\rule{0.277778em}{0ex}}x>{x}_{0}>0.$ 4

Gaussian, gamma, exponential and Weibull distributions, among others, are included in this family. An important part of our work consists in establishing links between models (2) and (4) in order to propose new estimation methods. We also consider the case where the observations were recorded with a covariate information. In this case, the extreme-value index and the ${p}_{n}$-quantile are functions of the covariate. We propose estimators of these functions by using moving window approaches, nearest neighbor methods, or kernel estimators.

### 3.3.2 Level sets estimation

Level sets estimation is a recurrent problem in statistics which is linked to outlier detection. In biology, one is interested in estimating reference curves, that is to say curves which bound $90%$ (for example) of the population. Points outside this bound are considered as outliers compared to the reference population. Level sets estimation can be looked at as a conditional quantile estimation problem which benefits from a non-parametric statistical framework. In particular, boundary estimation, arising in image segmentation as well as in supervised learning, is interpreted as an extreme level set estimation problem. Level sets estimation can also be formulated as a linear programming problem. In this context, estimates are sparse since they involve only a small fraction of the dataset, called the set of support vectors.

### 3.3.3 Dimension reduction

Our work on high dimensional data requires that we face the curse of dimensionality phenomenon. Indeed, the modelling of high dimensional data requires complex models and thus the estimation of high number of parameters compared to the sample size. In this framework, dimension reduction methods aim at replacing the original variables by a small number of linear combinations with as small as a possible loss of information. Principal Component Analysis (PCA) is the most widely used method to reduce dimension in data. However, standard linear PCA can be quite inefficient on image data where even simple image distortions can lead to highly non-linear data. Two directions are investigated. First, non-linear PCAs can be proposed, leading to semi-parametric dimension reduction methods 93. Another field of investigation is to take into account the application goal in the dimension reduction step. One of our approaches is therefore to develop new Gaussian models of high dimensional data for parametric inference 90. Such models can then be used in a Mixtures or Markov framework for classification purposes. Another approach consists in combining dimension reduction, regularization techniques, and regression techniques to improve the Sliced Inverse Regression method 94.

# 4 Application domains

## 4.1 Image Analysis

Participants: Florence Forbes, Jean-Baptiste Durand, Stephane Girard, Pedro Coelho Rodrigues, Benoit Kugler, Alexandre Constantin.

As regards applications, several areas of image analysis can be covered using the tools developed in the team. More specifically, in collaboration with team perception, we address various issues in computer vision involving Bayesian modelling and probabilistic clustering techniques. Other applications in medical imaging are natural. We work more specifically on MRI and functional MRI data, in collaboration with the Grenoble Institute of Neuroscience (GIN). We also consider other statistical 2D fields coming from other domains such as remote sensing, in collaboration with the Institut de Planétologie et d'Astrophysique de Grenoble (IPAG) and the Centre National d'Etudes Spatiales (CNES). In this context, we worked on hyperspectral and/or multitemporal images. In the context of the "pole de competivité" project I-VP, we worked of images of PC Boards.

## 4.2 Biology, Environment and Medicine

Participants: Florence Forbes, Stephane Girard, Jean-Baptiste Durand, Julyan Arbel, Sophie Achard, Pedro Coelho Rodrigues, Olivier Francois, Yuchen Bai, Theo Moins, Daria Bystrova, Meryem Bousebata, Lucrezia Carboni, Hana Lbath.

A third domain of applications concerns biology and medicine. We considered the use of mixture models to identify biomakers. We also investigated statistical tools for the analysis of fluorescence signals in molecular biology. Applications in neurosciences are also considered. In the environmental domain, we considered the modelling of high-impact weather events and the use of hyperspectral data as a new tool for quantitative ecology.

# 5 Social and environmental responsibility

## 5.1 Footprint of research activities

The footprint of our research activities has not been assessed yet. Most of the team members have validated the "charte d'éco-resposnsabilité" written by a working group from Laboratoire Jean Kuntzmann, which should have practical implications in the near future.

## 5.2 Impact of research results

A lot of our developments are motivated by and target applications in medicine and environmental sciences. As such they have a social impact with a better handling and treatment of patients, in particular with brain diseases or disorders. On the environmental side, our work has an impact on geoscience-related decision making with e.g. extreme events risk analysis, planetary science studies and tools to assess biodiversity markers. However, how to truly measure and report this impact in practice is another question we have not really addressed yet.

# 6 Highlights of the year

## 9.2 International research visitors

### 9.2.1 Visits of international scientists

Bernardo Nipoti, Bicocca University, Milan, Italy, visited the team in June for a collaboration with Julyan Arbel on Beta two-parameter processes in Bayesian nonparametrics.

### 9.2.2 Visits to international teams

Julyan Arbel visited the Casa Matemática Oaxaca (CMO), Mexico (November 28 - December 3) for a working group dedicated to bringing together researchers working on objective Bayesian methodology, Bayesian non-parametric methods, and machine learning. The Casa Matematica Oaxaca (CMO) in Mexico, and the Banff International Research Station for Mathematical Innovation and Discovery (BIRS) in Banff, Canada, are collaborative Canada-US-Mexico ventures that provide an environment for creative interaction as well as the exchange of ideas, knowledge, and methods within the mathematical sciences, with related disciplines and with industry. Other BIRS partners include the Institute for Advanced Study in Mathematics (Hangzhou, China) and the Institute of Mathematics at the University of Granada (Spain).

## 9.3 National initiatives

Participants: Jean-Baptiste Durand, Florence Forbes, Julyan Arbel, Sophie Achard, Stephane Girard, Alexandre Constantin, Meryem Bousebata, Giovanni Poggiato.

#### ANR

Statify is involved in the 4-year ANR project ExtremReg (2019-2023) hosted by Toulouse University. This research project aims to provide new adapted tools for nonparametric and semiparametric modeling from the perspective of extreme values. Our research program concentrates around three central themes. First, we contribute to the expanding literature on non-regular boundary regression where smoothness and shape constraints are imposed on the regression function and the regression errors are not assumed to be centred, but one-sided. Our second aim is to further investigate the study of the modern extreme value theory built on the use of asymmetric least squares instead of traditional quantiles and order statistics. Finally, we explore the less-discussed problem of estimating high-dimensional, conditional and joint extremes

The financial support for Statify is about 15.000 euros.

Statify is also involved in the ANR project GAMBAS (2019-2023) hosted by Cirad, Montpellier. The project Generating Advances in Modeling Biodiversity And ecosystem Services (GAMBAS) develops statistical improvements and ecological relevance of joint species distribution models. The project supports the PhD thesis of Giovanni Poggiato.

#### Grenoble Idex projects

Statify is involved in a cross-disciplinary project (CDP) Risk@UGA.

• The main objective of the Risk@UGA project is to provide some innovative tools both for the management of risk and crises in areas that are made vulnerable because of strong interdependencies between human, natural or technological hazards, in synergy with the conclusions of Sendai conference. The project federates a hundred researchers from Human and Social Sciences, Information & System Sciences, Geosciences and Engineering Sciences, already strongly involved in the problems of risk assessment and management, in particular natural risks. The PhD thesis of Meryem Bousebata is one of the eleven PhDs funded by this project.

In the context of the Idex associated with the Université Grenoble Alpes, Alexandre Constantin was awarded half a PhD funding from IRS (Initiatives de Recherche Stratégique), 50 keuros.

In the context of the MIAI (Multidisciplinary Institute in Artificial Intelligence) institute and its open call to sustain the development and promotion of AI, Stéphane Girard was awarded a grant of 4500 euros for his project "Simulation of extreme values by AI generative models. Application to banking risk" joint with CMAP, Ecole Polytechnique.

In the context of the MIAI (Multidisciplinary Institute in Artificial Intelligence) institute and its open call to sustain the development and promotion of AI, Julyan Arbel was awarded a grant of 5000 euros for his project "Bayesian deep learning".

Julyan Arbel was awarded a grant of 10000 euros for his project "Bayesian nonparametric modeling".

### 9.3.1 Networks

MSTGA and AIGM INRAE (French National Institute for Agricultural Research) networks: F. Forbes and J.B Durand are members of the INRAE network called AIGM (ex MSTGA) network since 2006, website, on Algorithmic issues for Inference in Graphical Models. It is funded by INRAE MIA and RNSC/ISC Paris. This network gathers researchers from different disciplines. Statify co-organized and hosted 2 of the network meetings in 2008 and 2015 in Grenoble.

# 10 Dissemination

Participants: Sophie Achard, Pedro Coelho Rodrigues, Florence Forbes, Julyan Arbel, Jean-Baptiste Durand, Stephane Girard, Olivier Francois, Hana Lbath.

## 10.1 Promoting scientific activities

### 10.1.1 Scientific events: organisation

#### General chair, scientific chair

• Brain connectivity networks: quality and reporducibility workshop in the context of the Complex Systems conference in October 2021 in Lyon. Details on the workshop website.

#### Member of the organizing committees

• Julyan Arbel was a member of the scientific and organizing committees of the French Mirror of the ISBA conference at CIRM in July 2021, of the Approximate Bayesian Computation meeting in April 2021, and of the Statistical Methods for Post Genomic Data analysis (SMPGD) meeting, in January 2021.

### 10.1.2 Scientific events: selection

#### Reviewer

• Julyan Arbel has been a rewiewer for the International Conference on Machine Learning (ICML), the Symposium on Advances in Approximate Bayesian Inference (AABI).
• Florence Forbes has been a reviewer for EUSIPCO 2021.

### 10.1.3 Journal

#### Member of the editorial boards

• Juyan Arbel is Associate Editor of Bayesian Analysis since 2019.
• Julyan Arbel and Florence Forbes are Associate Editors of Australian and New Zealand Journal of Statistics since 2019.
• Julyan Arbel is Associate Editor of Statistics & Probability Letters since 2019.
• Julyan Arbel is Associate Editor of Computational Statistics & Data Analysis since 2020.
• Stéphane Girard is Associate Editor of Dependence Modelling (De Gruyter) since 2015.
• Stéphane Girard is Associate Editor of Journal of Multivariate Analysis (Elsevier) depuis 2016.
• Stéphane Girard is Associate Editor of Revstat - Statistical Journal since 2019.

#### Reviewer - reviewing activities

• Julyan Arbel has been a rewiewer for Biometrika, Journal of Royal Statistical Society - series C, Journal of Multivariate Analysis, Scandinavian Journal of Statistics, IEEE Transactions on Signal Processing, Econometrics and Statistics, and for a book to be published in CRC Press.
• Stéphane Girard has been a rewiewer for JASA (Journal of the American Statistical Association) and EJS (Electronic Journal of Statistics).
• Florence Forbes has been a reviewer for Statistics & Computing, Bayesian Analysis, Australian and New Zealand journal of Statistics, Journal of Statistical Distributions and Applications.
• Pedro Rodrigues has been a reviewer for NeuroImage, IEEE Transactions on Biomedical Engineering, and Pattern Recognition.
• Jean-Baptiste Durand has been a reviewer for Ecology and Evolution and for Advances in Data Analysis and Classification.

### 10.1.4 Invited talks

• Florence Forbes has been invited as a plenary speaker to the international conference "End-to-end Bayesian Learning" at CIRM in October 2021, and to the workshop in honor of Christian Robert 60th birthday in September 2021.

F. Forbes was also invited to give a talk in special sessions at the ABC in Svalbard international conference and at the SIAM international conference on Uncertainty Quantification respectively in April and March 2021.

F. Forbes was then invited to give talks at the Laplace Daemon Criteo Online Seminar in March 2021, at the local Statistics department seminar at University of Grenoble and at the Grenoble Institute of Neuroscience neuroimaging team days in September 2021.

• Julyan Arbel has been invited to give a talk at the OxCSML Seminar at University of Oxford in April 2021, at the ApproxBayes team Seminar, RIKEN AIP, Tokyo, Japan in May, at Journées MAS (Modélisation Aléatoire et Statistique), France, in August, and at the Foundations of Objective Bayesian Methodology Workshop, Casa Matemática Oaxaca (CMO), Mexico, in December.
• Stéphane Girard was an invited speaker at the 14th International Conference of the ERCIM WG on Computing and Statistics 50 and the 13th International Workshop on Rare-Event Simulation 51.
• Pedro Rodrigues was invited to give a talk to the GAIA team from GIPSA-lab in December 2021.

### 10.1.5 Scientific expertise

• Florence Forbes has reviewed projects for the Swiss Personalized Health and Related Technologies (PHRT) Pioneer Imaging Projects.
• Florence Forbes is a member of the Helmholtz AI Cooperation Unit advisory committee, 2019-present.
• Florence Forbes and Sophie Achard are members of the EURASIP Technical Area Committee BISA (Biomedical Image & Signal Analytics) since January 2021 for a 3 years duration.
• Florence Forbes was a member of committees in charge of hiring professors and teaching assistants at Ecole Polytechnique, Paris, and at the Agricultural Science School in Rennes (AgroCampus ouest), and of the committee in charge of hiring Inria junior researchers for the Grenoble center.
• Stéphane Girard was a reviewer for the Hi!Paris Fellowships program 2021 Call.

• Sophie Achard, since Nov. 2020, has been elected as the head of Pole MSTIC (with Jean-Paul Jamont, Karine Altisen and Christine Lescop) at University of Grenoble.
• Florence Forbes is since July 2021 Deputy head of science (DSA) for the Inria Grenoble center.
• Julyan Arbel is a member of the scientific committee of the Data Science axis of Persyval Labex.

## 10.2 Teaching - Supervision - Juries

### 10.2.1 Teaching

• Master : Stéphane Girard, Statistique Inférentielle Avancée, 18 ETD, M1 level, Ensimag. Grenoble-INP, France.
• Master : Stéphane Girard, Introduction to Extreme-Value Analysis, 27 ETD, M2 level, Univ-Grenoble Alpes (UGA), France.
• Master and PhD course: Julyan Arbel, Bayesian nonparametrics and Bayesian deep learning, Master Mathématiques Apprentissage et Sciences Humaines (M*A*S*H), Université PSL (Paris Sciences & Lettres), 25 ETD. Bayesian deep learning, Master Intelligence Artificielle, Systèmes, Données (IASD), Université PSL (Paris Sciences & Lettres), 12 ETD.
• Master and PhD course: Julyan Arbel, Bayesian machine learning, Master Mathématiques Vision et Apprentissage Master MVA, École normale supérieure Paris-Saclay, 36 ETD.
• Master: Jean-Baptiste Durand, Statistics and probability, 192H, M1 and M2 levels, Ensimag Grenoble INP, France. Head of the MSIAM M2 program, in charge of the data science track.
• Jean-Baptiste Durand is a faculty member at Ensimag, Grenoble INP.
• Sophie Achard M1 course Théorie des graphes et réseaux sociaux, M1 level, MIASHS, Université Grenoble Alpes (UGA), 14 ETD.

### 10.2.2 Supervision

PhD Defended: Benoit Kugler, "Massive hyperspectral images analysis by inverse regression of physical models", Florence Forbes and Sylvain Douté, Université Grenoble Alpes, Defended in July 2021.

PhD Defended: Alexandre Constantin "Analyse de séries temporelles massives d'images satellitaires : Applications à la cartographie des écosystèmes", Stéphane Girard and Mathieu Fauvel, Université Grenoble Alpes, Defended in December 2021.

PhD in progress: Mariia Vladimirova, “Prior specification for Bayesian deep learning models and regularization implications”, started on October 2018, Julyan Arbel, Université Grenoble Alpes.

PhD in progress: Meryem Bousebata "Bayesian estimation of extreme risk measures: Implication for the insurance of natural disasters", started on October 2018, Stéphane Girard and Geffroy Enjolras, Université Grenoble Alpes.

PhD in progress: Daria Bystrova, “Joint Species Distribution Modeling: Dimension reduction using Bayesian nonparametric priors”, started on October 2019, Julyan Arbel and Wilfried Thuiller, Université Grenoble Alpes.

PhD in progress: Giovanni Poggiatto, “Scalable Approaches for Joint Species Distribution Modeling”, started on November 2019, Julyan Arbel and Wilfried Thuiller, Université Grenoble Alpes.

PhD in progress: Théo Moins "Quantification bayésienne des limites d’extrapolation en statistique des valeurs extrêmes", started on October 2020, Stéphane Girard and Julyan Arbel, Université Grenoble Alpes.

PhD in progress: Michael Allouche "Simulation d’extrêmes par modèles génératifs et applications aux risques bancaires", started on April 2020, Stéphane Girard and Emmanuel Gobet, Ecole Polytechnique.

PhD in progress: Minh Tri Lê, “Constrained signal processing using deep neural networks for MEMs sensors based applications.”, started on September 2020, Julyan Arbel and Etienne de Foras, Université Grenoble Alpes, CIFRE Invensense.

PhD in progress: Hana Lbath, "Advanced Spatiotemporal Statistical Models for Quantification and Estimation of Functional Connectivity", started in October 2020, supervised by Sophie Achard and Alex Petersen (Brigham Young University, Utah, USA).

PhD in progress: Lucrezia Carboni, "Graph embedding for brain connectivity", started in October 2020, supervised by Sophie Achard and Michel Dojat (GIN).

PhD in progress: Louise Alamichel. "Bayesian Nonparametric methods for complex genomic data" Inria, started in October 2021, advised by Julyan Arbel and Guillaume Kon Kam King (INRAE).

PhD in progress: Yuchen Bai, "Hierarchical Bayesian Modelling of leaf area density from UAV-lidar", started in October 2021, supervised by Jean-Baptiste Durand, Florence Forbes and Gregoire Vincent (IRD, Montpellier).

PhD in progress: Julia Linhart, "Simulation based inference with neural networks: applications to computational neuroscience", started in November 2021, supervised by Pedro Rodrigues and Alexandre Gramfort (DR Inria Saclay).

### 10.2.3 Juries

• Florence Forbes has been a reviewer for the HDR of Pierre Maurel (Rennes).
• Florence Forbes has been a member of the PhD committees of Hamza Cherkaoui (Saclay), Amelie Barbe (Lyon); and chair for the commitees of Olga Permiakova (Grenoble) and Remi Souriau (Saclay).
• Florence Forbes has been a member of the intermediate PhD committee of Nicolas Pinon (Lyon) and reviewer for the Master commitee of Michael Carr (QUT Brisbane Australia).
• Stéphane Girard has been a member of the PhD committee of Benoit Colange (Université Savoie-Mont Blanc).
• Stéphane Girard has been a member of the intermediate PhD committee of Valentine Bellet and Erwan Giry-Fouquet (Université de Toulouse).

### 10.2.4 Articles and contents

Sophie Achard and Jean-Baptiste Durand published an illustration of the process of statistical modeling, in the Inria journal for popularization of science (in French). The topic was illustrated through the analysis of eye movements to infer cognitive processes (see Subsection 7.3.8).

# 11 Scientific production

## 11.1 Major publications

• 1 articleC.C. Amblard and S.S. Girard. Estimation procedures for a semiparametric family of bivariate copulas.Journal of Computational and Graphical Statistics1422005, 1--15
• 2 articleJ.J. Blanchet and F.F. Forbes. Triplet Markov fields for the supervised classification of complex structure data.IEEE trans. on Pattern Analyis and Machine Intelligence30(6)2008, 1055--1067
• 3 articleC.C. Bouveyron, S.S. Girard and C.C. Schmid. High dimensional data clustering.Computational Statistics and Data Analysis522007, 502--519
• 4 articleC.C. Bouveyron, S.S. Girard and C.C. Schmid. High dimensional discriminant analysis.Communication in Statistics - Theory and Methods36142007
• 5 articleF.Fabien Boux, F.Florence Forbes, J.Julyan Arbel, B.Benjamin Lemasson and E. L.Emmanuel L. Barbier. Bayesian inverse regression for vascular magnetic resonance fingerprinting.IEEE Transactions on Medical Imaging407July 2021, 1827-1837
• 6 articleA.Abdelaati Daouia, S.Stéphane Girard and G.G. Stupfler. Estimation of Tail Risk based on Extreme Expectiles.Journal of the Royal Statistical Society series B802018, 263--292
• 7 articleA.Antoine Deleforge, F.Florence Forbes and R.Radu Horaud. High-Dimensional Regression with Gaussian Mixtures and Partially-Latent Response Variables.Statistics and ComputingFebruary 2014
• 8 articleF.Florence Forbes and G.G. Fort. Combining Monte Carlo and Mean field like methods for inference in hidden Markov Random Fields.IEEE trans. Image Processing1632007, 824-837
• 9 articleF.Florence Forbes and D.Darren Wraith. A new family of multivariate heavy-tailed distributions with variable marginal amounts of tailweights: Application to robust clustering.Statistics and Computing246November 2014, 971-984
• 10 articleS.S. Girard. A Hill type estimate of the Weibull tail-coefficient.Communication in Statistics - Theory and Methods3322004, 205--234
• 11 articleH.Hongliang Lu, J.Julyan Arbel and F.Florence Forbes. Bayesian nonparametric priors for hidden Markov random fields.Statistics and Computing302020, 1015-1035

## 11.2 Publications of the year

### International journals

• 12 articleJ.Julyan Arbel, R.Riccardo Corradin and B.Bernardo Nipoti. Dirichlet process mixtures under affine transformations of the data.Computational Statistics36March 2021, 577-601
• 13 articleF.Fabien Boux, F.Florence Forbes, J.Julyan Arbel, B.Benjamin Lemasson and E. L.Emmanuel L. Barbier. Bayesian inverse regression for vascular magnetic resonance fingerprinting.IEEE Transactions on Medical Imaging407July 2021, 1827-1837
• 14 articleF.Fabien Boux, F.Florence Forbes, N.Nora Collomb, E.Emma Zub, L.Lucile Maziere, F.Fréderic Bock, M.Marine Blaquière, V.Vasile Stupar, A.Antoine Depaulis, N.Nicola Marchi and E.Emmanuel Barbier. Neurovascular multiparametric MRI defines epileptogenic and seizure propagation regions in experimental mesiotemporal lobe epilepsy.Epilepsia625May 2021, 1244-1255
• 15 articleD.Daria Bystrova, G.Giovanni Poggiato, B.Billur Bektaş, J.Julyan Arbel, J. S.James S Clark, A.Alessandra Guglielmi and W.Wilfried Thuiller. Clustering species with residual covariance matrix in Joint Species Distribution models.Frontiers in Ecology and Evolution9March 2021, 601384:1-11
• 16 articleJoint Supervised Classification and Reconstruction of Irregularly Sampled Satellite Image Times Series.IEEE Transactions on Geoscience and Remote Sensing60May 2021, 4403913
• 17 articleExpectHill estimation, extreme risk and heavy tails.Journal of Econometrics2211March 2021, 97-117
• 18 articleL.Laurent Gardes and S.Stéphane Girard. On the estimation of the variability in the distribution tail.Test30December 2021, 884--907
• 19 articleS.Stéphane Girard, H.Hadrien Lorenzo and J.Jérôme Saracco. Advanced topics in Sliced Inverse Regression.Journal of Multivariate Analysis1882022, 104852
• 20 articleS.Stéphane Girard, G. C.Gilles Claude Stupfler and A.Antoine Usseglio-Carleve. Extreme Conditional Expectile Estimation in Heavy-Tailed Heteroscedastic Regression Models.Annals of Statistics496December 2021, 3358--3382
• 21 articleFunctional estimation of extreme conditional expectiles.Econometrics and Statistics 21January 2022, 131-158
• 22 articleNonparametric extreme conditional expectile estimation.Scandinavian Journal of Statistics491March 2022, 78-115
• 23 articleB.Benoit Kugler, F.Florence Forbes and S.Sylvain Douté. Fast Bayesian Inversion for high dimensional inverse problems.Statistics and Computing2021
• 24 articleT.Thomas Mistral, P.Pauline Roca, C.Christophe Maggia, A.Alan Tucholka, F.Florence Forbes, S.Senan Doyle, A.Alexandre Krainik, D.Damien Galanaud, E.Emmanuelle Schmitt, S.Stéphane Kremer, A.Adrian Kastler, I.Irène Troprès, E.Emmanuel Barbier, J.-F.Jean-François Payen and M.Michel Dojat. Automatic quantification of brain lesion volume from post-trauma MR Images.Frontiers in Neurology2021
• 25 articleT.Théo Moins, J.Julyan Arbel, A.Anne Dutfoy and S.Stéphane Girard. Discussion of the paper "Rank-Normalization, Folding, and Localization: An Improved $\stackrel{}{R}$ for Assessing Convergence of MCMC”.Bayesian Analysis1622021, 711--712
• 26 articleH. D.Hien Duy Nguyen and F.Florence Forbes. Global implicit function theorems and the online Expectation-Maximisation algorithm.Australian and New Zealand Journal of Statistics2022, 1-27
• 27 articleSplitting models for multivariate count data.Journal of Multivariate Analysis181January 2021, 104677
• 28 articleG.Giovanni Poggiato, T.Tamara Münkemüller, D.Daria Bystrova, J.Julyan Arbel, J.James Clark and W.Wilfried Thuiller. On the interpretations of joint modelling in community ecology.Trends in Ecology and Evolution365May 2021, 391-401
• 29 articleS. M.Sandra M. Potin, S.Sylvain Douté, B.Benoit Kugler and F.Florence Forbes. The impact of asteroid shapes and topographies on their reflectance spectroscopy.Icarus376April 2022, 114806:1-24
• 30 articleF.Félix Renard, C.Christian Heinrich, M.Marine Bouthillon, M.Maleka Schenck, F.Francis Schneider, S.Stéphane Kremer and S.Sophie Achard. A covariate-constraint method to map brain feature space into lower dimensional manifolds.Network Neuroscience51January 2021, 252-273

### International peer-reviewed conferences

• 31 inproceedingsD.Daria Bystrova, J.Julyan Arbel, G.Guillaume Kon Kam King and F.François Deslandes. Approximating the clusters' prior distribution in Bayesian nonparametric models.AABI 2020 - 3rd Symposium on Advances in Approximate Bayesian InferenceOnline, United StatesJanuary 2021, 1-16
• 32 inproceedingsL.Lucrezia Carboni, S.Sophie Achard and M.Michel Dojat. Network embedding for brain connectivity.ISBI 2021 - International Symposium on Biomedical ImagingNice / Virtual, FranceApril 2021, 1-4
• 33 inproceedingsB.Benjamin Lambert, M.Maxime Louis, S.Senan Doyle, F.Florence Forbes, M.Michel Dojat and A.Alan Tucholka. Leveraging 3D information in unsupervised brain MRI segmentation.ISBI 2021 - 18th International Symposium on Biomedical ImagingNice / Virtual, FranceApril 2021, 1-4
• 34 inproceedingsH.Hanâ Lbath, A.Angela Bonifati and R.Russ Harmer. Schema Inference for Property Graphs.EDBT 2021 - 24th International Conference on Extending Database TechnologyEDBTNicosia, CyprusMarch 2021, 499-504
• 35 inproceedingsV.Verónica Muñoz-Ramírez, N.Nicolas Pinon, F.Florence Forbes, C.Carole Lartizien and M.Michel Dojat. Patch vs. Global Image-Based Unsupervised Anomaly Detection in MR Brain Scans of Early Parkinsonian Patients.MLCN 2021 - 4th International Workshop in Machine Learning in Clinical Neuroimaging13001Lecture Notes in Computer ScienceStrasbourg, FranceSpringer International PublishingSeptember 2021, 34-43
• 36 inproceedingsM.Mariia Vladimirova, J.Julyan Arbel and S.Stéphane Girard. Dependence between Bayesian neural network units.BDL 2021 - Workshop. Bayesian Deep Learning NeurIPSMontreal, CanadaDecember 2021, 1-9

### National peer-reviewed Conferences

• 37 inproceedingsM.Michaël Allouche, S.Stéphane Girard and E.Emmanuel Gobet. On the approximation of extreme quantiles with neural networks.SFdS 2021 - 52èmes Journées de Statistique de la Société Française de StatistiqueNice, FranceJune 2021, 1-5
• 38 inproceedingsM.Meryem Bousebata, G.Geoffroy Enjolras and S.Stéphane Girard. Single-index Extreme-PLS regression.JDS 2021 - 52èmes Journées de Statistique organisées par la Société Française de Statistique (SFdS)Nice / Virtual, FranceJune 2021, 1-6
• 39 inproceedingsT.Théo Moins, J.Julyan Arbel, A.Anne Dutfoy and S.Stéphane Girard. On Reparameterisations of the Poisson Process Model for Extremes in a Bayesian Framework.JDS 2021 - 52èmes Journées de Statistique de la Société Française de Statistique (SFdS)Nice / Virtual, FranceJune 2021, 1-6

### Conferences without proceedings

• 40 inproceedingsM.Michaël Allouche, S.Stéphane Girard and E.Emmanuel Gobet. Generative model for fbm with deep ReLU neural networks.Bernoulli-IMS 2021 - 10th World Congress in Probability and StatisticsSeoul / Virtual, South KoreaJuly 2021
• 41 inproceedingsM.Michaël Allouche, S.Stéphane Girard and E.Emmanuel Gobet. On the approximation of extreme quantiles with ReLU neural networks.EVA 2021 - 12th International Conference on Extreme Value AnalysisEdinburgh / Virtual, United KingdomJune 2021
• 42 inproceedingsJ.Julyan Arbel, M.Mario Beraha and D.Daria Bystrova. Bayesian block-diagonal graphical models via the Fiedler prior.SFdS - 52 Journées de Statistique de la Société Francaise de StatistiqueNice, FranceJune 2021, 1-6
• 43 inproceedingsJ.Julyan Arbel, F.Florence Forbes, H. D.Hien Duy Nguyen and T.TrungTin Nguyen. Approximate Bayesian computation with surrogate posteriors.ISBA 2021 - World Meeting of the International Society for Bayesian AnalysisMarseille, FranceJune 2021
• 44 inproceedingsJ.Julyan Arbel, S.Stéphane Girard, T.Théo Moins, A.Anne Dutfoy and K.Khalil Leachouri. Improving MCMC convergence diagnostic with a local version of R-hat.MAS 2021 - Journées Modélisation Aléatoire et StatistiqueOrléans, FranceAugust 2021
• 45 inproceedingsM.Meryem Bousebata, G.Geoffroy Enjolras and S.Stéphane Girard. Extreme partial least-squares regression.EVA 2021 - 12th International Conference on Extreme Value AnalysisEdinburgh / Virtual, United KingdomJune 2021
• 46 inproceedingsM.Meryem Bousebata, G.Geoffroy Enjolras and S.Stéphane Girard. Extreme partial least-squares regression.CMStatistics 2021 - 14th International Conference of the ERCIM WG on Computational and Methodological StatisticsLondon, United KingdomDecember 2021
• 47 inproceedingsT.Thomas Coudert, S.Sophie Ancelet, N.Nadya Pyatigorskaya, L.Lucia Nichelli, D.Damien Ricard, D.Dimitri Psimaras, M. O.Marie Odile Bernier, M.Michel Dojat, F.Florence Forbes and A.Alan Tucholka. Apport du Transfer Learning pour la segmentation automatique de lésions cérébrales radio-induites chez des patients atteints de glioblastome à partir d’un nombre restreint d’IRMs annotées.Journées de biostatistique 2021 du GDR « Statistiques & Santé »virtuel, France2021
• 48 inproceedingsJ.Jonathan El Methni and S.Stéphane Girard. A bias-reduced version of the Weissman estimator for extreme value-at-risk.CMStatistics 2021 - 14th International Conference of the ERCIM WG on Computational and Methodological StatisticsLondon, United KingdomDecember 2021
• 49 inproceedingsJ.Jonathan El Methni and S.Stéphane Girard. A bias-reduced version of the Weissman extreme quantile estimator.EVA 2021 - 12th International Conference on Extreme Value AnalysisEdinburgh / Virtual, United KingdomJune 2021
• 50 inproceedingsEstimation of the largest tail-index and extreme quantiles from a mixture of heavy-tailed distributions.CMStatistics 2021 - 14th International Conference of the ERCIM WG on Computational and Methodological StatisticsLondon, United KingdomDecember 2021
• 51 inproceedingsEstimation of the tail-index and extreme quantiles from a mixture of heavy-tailed distributions.RESIM 2021 - 13th International Workshop on Rare-Event SimulationParis / Virtual, FranceMay 2021, 1
• 52 inproceedingsB.Benoit Kugler, F.Florence Forbes and S.Sylvain Douté. First order Sobol indices for physical models via inverse regression.JDS 2020 - 52èmes Journées de Statistique de la Société Française de Statistique (SFdS)Nice, FranceJune 2021, 1-6
• 53 inproceedingsB.Benoit Kugler, F.Florence Forbes, S.Sylvain Douté and M.Michel Gay. Efficient Bayesian data assimilation via inverse regression.SFdS 2020 - 52èmes Journées de Statistiques de la Société Française de StatistiqueNice, FranceJune 2021, 1-6
• 54 inproceedingsB.Benjamin Lambert, F.Florence Forbes, S.Senan Doyle, A.Alan Tucholka and M.Michel Dojat. Fast Uncertainty Quantification for Deep Learning-based MR Brain Segmentation.EGC 2022 - Conference francophone pour l'Extraction et la Gestion des ConnaissancesBlois, FranceJanuary 2022, 1-12
• 55 inproceedingsB.Benjamin Lambert, F.Florence Forbes, A.Alan Tucholka, S.Senan Doyle and M.Michel Dojat. Multi-Scale Evaluation of Uncertainty Quantification Techniques for Deep Learning based MRI Segmentation.ISMRM-ESMRMB & ISMRT 2022 - 31st Joint Annual Meeting International Society for Magnetic Resonance in MedecineLondon, United KingdomMay 2022, 1-3
• 56 inproceedingsT.Théo Moins, J.Julyan Arbel, A.Anne Dutfoy and S.Stéphane Girard. A Bayesian Framework for Poisson Process Characterization of Extremes with Objective Prior.ISBA 2021 - World Meeting of the International Society for Bayesian AnalysisVirtual, FranceJune 2021
• 57 inproceedingsT.Théo Moins, J.Julyan Arbel, A.Anne Dutfoy and S.Stéphane Girard. A Bayesian framework for Poisson process characterization of extremes with uninformative prior.CMStatistics 2021 - 14th International Conference of the ERCIM WG on Computational and Methodological StatisticsLondon, United KingdomDecember 2021
• 58 inproceedingsA.Antoine Usseglio-Carleve, S.Stéphane Girard and G.Gilles Stupfler. Extreme conditional expectile estimation in heavy-tailed heteroscedastic regression models.CMStatistics 2021 - 14th International Conference of the ERCIM WG on Computational and Methodological StatisticsLondon, United KingdomDecember 2021
• 59 inproceedingsExtreme expectile regression: theory and applications.EVA 2021 - 12th International Conference on Extreme Value AnalysisEdinburgh / Virtual, United KingdomJune 2021
• 60 inproceedingsBayesian neural network unit priors and generalized Weibull-tail property.ACML 2021 - 13th Asian Conference on Machine LearningVirtual, Unknown RegionNovember 2021, 1-16
• 61 inproceedingsM.Mariia Vladimirova, J.Julyan Arbel and S.Stéphane Girard. Generalized Weibull-tail distributions.JDS 2021 - 52èmes Journées de Statistique de la Société Française de Statistique (SFdS)Nice, FranceJune 2021, 1-6

### Scientific book chapters

• 62 inbookS.Stéphane Girard, G.Gilles Stupfler and A.Antoine Usseglio-Carleve. Extreme Lp-quantile kernel regression.Advances in Contemporary Statistics and EconometricsSpringer2021, 197-219

### Doctoral dissertations and habilitation theses

• 63 thesisB.Benoit Kugler. Massive hyperspectral images analysis by inverse regression of physical models.Université Grenoble Alpes [2020-....]July 2021

### Reports & preprints

• 64 miscWhittle estimation with (quasi-)analytic wavelets.June 2021
• 65 miscM.Michaël Allouche, J.Jonathan El Methni and S.Stéphane Girard. A refined Weissman estimator for extreme quantiles.March 2022
• 66 miscM.Michaël Allouche, S.Stéphane Girard and E.Emmanuel Gobet. A generative model for fBm with deep ReLU neural networks.January 2022
• 67 miscM.Michaël Allouche, S.Stéphane Girard and E.Emmanuel Gobet. EV-GAN: Simulation of extreme events with ReLU neural networks.June 2021
• 68 miscJ.Julyan Arbel, S.Stéphane Girard, H. D.Hien Duy Nguyen and A.Antoine Usseglio-Carleve. Multivariate expectile-based distribution: properties, Bayesian inference, and applications.November 2021
• 69 miscK.Karina Ashurbekova, A.Antoine Usseglio-Carleve, F.Florence Forbes and S.Sophie Achard. Optimal shrinkage for robust covariance matrix estimators in a small sample size setting.March 2021
• 70 miscM.Meryem Bousebata, G.Geoffroy Enjolras and S.Stéphane Girard. Extreme Partial Least-Squares regression.2021
• 71 miscD.Daria Bystrova, J.Julyan Arbel and T.Thibaud Rahier. Contributed comment on Article by Hahn, Murray, and Carvalho.August 2020
• 72 miscD.Daria Bystrova, G.Giovanni Poggiato, J.Julyan Arbel and W.Wilfried Thuiller. Latent factor models: a tool for dimension reduction in joint species distribution models.February 2021
• 73 miscMixture of multivariate gaussian processes for classification of irregularly sampled satellite image time-series.2021
• 74 reportP. A.Pascal Alain Dkengne Sielenou and S.Stéphane Girard. Estimation of upper bounds for return levels: Methodology and illustrations on univariate random sequences.Inria - Research Centre Grenoble – Rhône-AlpesFebruary 2022
• 75 miscJ.-B.Jean-Baptiste Durand, F.Florence Forbes, C. D.Cong Duc Phan, L.Long Truong, H. D.Hien D Nguyen and F.Fatoumata Dama. Bayesian nonparametric spatial prior for traffic crash risk mapping: a case study of Victoria, Australia.February 2021
• 76 miscF.Florence Forbes, H. D.Hien Duy Nguyen, T. T.Trung Tin Nguyen and J.Julyan Arbel. Approximate Bayesian computation with surrogate posteriors.February 2021
• 77 miscOn automatic bias reduction for extreme expectile estimation.January 2021
• 78 miscM. N.Mahouton Norbert Hounkonnou and P. D.Pascal Dkengne Sielenou. Conservation laws for under determined systems of differential equations.February 2022
• 79 miscN.Norbert Mahouton and P. D.Pascal Dkengne Sielenou. Extremum conditions for functionals involving higher derivatives of several variable vector valued functions.February 2022
• 80 miscV.Vincent Miele, C.Catherine Matias, M.Marc Ohlmann, G.Giovanni Poggiato, S.Stéphane Dray and W.Wilfried Thuiller. Quantifying the overall effect of biotic interactions on species communities along environmental gradients.March 2021
• 81 miscT.Théo Moins, J.Julyan Arbel, A.Anne Dutfoy and S.Stéphane Girard. On the use of a local $\stackrel{}{R}$ to improve MCMC convergence diagnostic.March 2022
• 82 miscT.TrungTin Nguyen, F.Faicel Chamroukhi, H. D.Hien Duy Nguyen and F.Florence Forbes. A non-asymptotic model selection in block-diagonal mixture of polynomial experts models.May 2021
• 83 miscH. D.Hien Duy Nguyen, F.Florence Forbes, G.Gersende Fort and O.Olivier Cappé. An online Minorization-Maximization algorithm.January 2022
• 84 miscT.TrungTin Nguyen, H. D.Hien Duy Nguyen, F.Faicel Chamroukhi and F.Florence Forbes. A non-asymptotic penalization criterion for model selection in mixture of experts models.May 2021
• 85 reportB.Brice Olivier, A.Anne Guérin-Dugué and J.-B.Jean-Baptiste Durand. Hidden Semi-Markov Models to Segment Reading Phases from Eye Movements.RR-9398Inria Grenoble - Rhône-AlpesMarch 2021

### Other scientific publications

• 86 inproceedingsA.Aurélien Delphin, F.Fabien Boux, C.Clément Brossard, J. M.Jan M Warnking, B.Benjamin Lemasson, E. L.Emmanuel L Barbier and T.Thomas Christen. Optimisation des patterns de signaux pour l'IRM Fingerprinting Vasculaire.SFRMBM 2021 - 5ème Congrès scientifique de la Société Française de Résonance Magnétique en Biologie et MédecineLyon, FranceSeptember 2021, 1-1
• 87 inproceedingsT.Théo Moins, J.Julyan Arbel, S.Stéphane Girard and A.Anne Dutfoy. Improving MCMC convergence diagnostic: a local version of R-hat.BayesComp-ISBA workshop: Measuring the quality of MCMC outputonline, FranceOctober 2021
• 88 inproceedingsS.Sandra Potin Manigand, S.Sylvain Douté, B.Benoit Kugler and F.Florence Forbes. Comparison of photometric phase curves resulting from various observation scenes.EPSC 2021 - Europlanet Science CongressGrenoble, FranceSeptember 2021, 1-1

## 11.3 Other

### Scientific popularization

• 89 articleComprendre un processus cognitif grâce à l’analyse statistique du mouvement des yeux.IntersticesDecember 2021

## 11.4 Cited publications

• 90 phdthesisC.C. Bouveyron. Modélisation et classification des données de grande dimension. Application à l'analyse d'images.Université Grenoble 1septembre 2006,
• 91 bookP.P. Embrechts, C.C. Klüppelberg and T.T. Mikosh. Modelling Extremal Events.33Applications of MathematicsSpringer-Verlag1997
• 92 bookF.F. Ferraty and P.P. Vieu. Nonparametric Functional Data Analysis: Theory and Practice.Springer Series in Statistics, Springer2006
• 93 phdthesisS.S. Girard. Construction et apprentissage statistique de modèles auto-associatifs non-linéaires. Application à l'identification d'objets déformables en radiographie. Modélisation et classification.Université de Cery-Pontoiseoctobre 1996
• 94 articleK.K.C. Li. Sliced inverse regression for dimension reduction.Journal of the American Statistical Association861991, 316--327
• 95 articleJ.J. Simola, J.J. Salojärvi and I.I. Kojo. Using hidden Markov model to uncover processing states from eye movements in information search tasks.Cognitive Systems Research94Oct 2008, 237-251