Science in industrial and applied mathematics (MSIAM)

Content

This course relies on practical sessions.

Visit the full page about this course

Contents:

• Types of equations, conservation laws
• Finite differences methods
• Laplace equation
• Parabolic equations (diffusion)
• Hyperbolic equations (propagation)
• Non linear hyperbolic equations

This course include practical sessions.

This is a two parts course, this part is the course mutualized with Ensimag 2A 4MMMEDPS.

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Contents:

• Image definition
• Fourier transform, FFT, applications
• Image digitalisation, sampling
• Image processing: convolution, filtering. Applications
• Image decomposition, multiresolution. Application to compression

This course includes practical sessions.

Visit the full page about this course

This course is an introduction to the differential geometry of curves and surfaces with a particular focus on spline curves and surfaces that are routinely used in geometrical design softwares.

Content

This course includes practical sessions.

Visit the full page about this course

The syllabus of the English course in M2 aims at enabling students to validate 3 competences that will be essential for their working life or for their doctoral studies in the future, at the B2 level:

1°) CAN give a clear presentation on a familiar topic, and CAN answer predictable or factual questions

2°) CAN find relevant information and essential points in written texts

3°) CAN make simple notes that are of reasonable use for essay or revision purposes, capturing most important points." (CEF, appendix D).

Reading comprehension can be validated in M1.

The course contents are linked to the students' fields of studies.

Visit the full page about this course

Contents:

This course includes practical sessions.

Visit the full page about this course

Ce cours présente des manières géométriques de traiter et résoudre des problèmes décrits par des équations différentielles.

        - chapitre I : Introduction : généralités sur les systèmes dynamiques

        - chapitre II : Systèmes unidimensionnels : Les points fixes, linéarisation et stabilité, Exemple : le modèle logistique, Existence et unicité des solutions d'équations différentielles ordinaires

        - chapitre III :  Bifurcations : Bifurcation selle-nœud, Bifurcation transcritique, Bifurcation transcritique imparfaite, Bifurcation fourche, Bifurcation fourche supercritique, Bifurcation fourche sous-critique, Bifurcation fourche supercritique imparfaite

        - chapitre IV :  Champ de vecteur sur un cercle : Oscillateur uniforme, Oscillateur non-uniforme

        - chapitre V : Flots bidimensionnels et applications : Existence et unicité des solutions et conséquences topologiques, Systèmes linéaires, Systèmes non-linéaires : linéarisation proche des points fixes, Cycles limites, Le théorème de Poincaré-Bendixson, Systèmes Liénard, Systèmes gradients, Fonctions de Liapunov

        - chapitre VI : Bifurcations bidimensionnelles : Bifurcations selle-nœud, transcritique et fourche, Bifurcation de Hopf, Bifurcations globales de cycles

Visit the full page about this course

This course presents basic notions on instabilities and turbulence. We try to be as progressive as possible and to base our presentation on analyses of real experiments and real flows. We review few mathematical methods to analyze nonlinear systems in terms of instabilities. The students have to use their new knowledge to run and analyze numerical simulations of very simple systems. We then study some of the most important physical mechanisms for fluid instabilities and the corresponding criteria. We quickly present a zoology of common fluid instabilities and discuss the mechanisms and the possible technical implications. We give a broad introduction on turbulence and describe few fundamental methods and results, in particular the Richardson cascade, the Reynolds decomposition and the Kolmogorov spectra.

Teaching program:

1. General introduction

- Instabilities and turbulence, interest?

- Reynolds experiment and Reynolds number

- Incompressible Navier-Stokes equations: diffusion and advection

- An example: the wake of a cylinder

2. Instabilities and transition to turbulence

- Systems with few degrees of freedom

- Fluid instability mechanisms and conditions

- Other flows examples

3. Effects of variable density

- Boussinesq approximation

- Unstable stratification, Rayleigh-Taylor instability

- Rayleigh-Benard instability (Ra, Nu)

- Stable stratification, Kelvin-Helmoltz instability and Richardson number

4. Turbulence

- Introduction, Richardson cascade

- Average and Reynolds decomposition

- Experimental and numerical methods to study turbulence

- Statistical descriptions

For this course, the students have to write in LaTeX a report on their practical work. Thus, we spent some time for a first gentle introduction of this tool widely used in scientific academia.

Visit the full page about this course

Ce module est une introduction à la turbulence phénoménologique et statistique. On s’intéresse aux définitions et propriétés de la turbulence en terme de processus physiques et leur description dans des familles types d’écoulements cisaillés que l’on peut retrouver dans la nature et en ingénierie.

jet turbulent
jet turbulent

Visit the full page about this course

Contents:

• Introduction to database
• Introduction to big data
• Introduction to high performance computing (HPC)
• Numerical solvers for HPC

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

January science and/or industrial project.

Visit the full page about this course

Industrial and/or research internship.

Visit the full page about this course

This program combines case studies coming from real life problems or models and lectures providing the mathematical and numerical backgrounds.

Contents:

• Introduction, classification, examples.
• Theoretical results: convexity and compacity, optimality conditions, KT theorem
• Algorithmic for unconstrained optimisation (descent, line search, (quasi) Newton)
• Algorithms for non differentiable problems
• Algorithms for constrained optimisation: penalisatio, SQP methods
• Applications

Visit the full page about this course

Visit the full page about this course

Operational Research proposes scientific methods
to help make better decisions. The idea is to develop and use mathematical and computer
tools to master complex problems. Practical applications are historically in
the direction and management of large systems of people, machines, and
materials in industry, service, humanitarian aid, environment ...

In this course, we will focus on especially on problems with a combinatorial
structure: the number of possible solutions is finite but too large to be
enumerated. The study of these problems involves a phase of modelling practical
problems and then algorithmic resolution.

At the end of this course, students will be able to propose a model and will be
able to implement practical solutions (dedicated or industrial tools) to deal
with a problem of decision or optimization.

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

To acquire the main theoretical and practical notions of modern cryptography: from notions in algorithmic complexity and information theory, to a general overview on the main algorithms and protocols in symmetric and asymmetric cryptography.

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Computer Graphics covers the set of techniques enabling the synthesis of animated virtual worlds. The applications range from entertainment (special effects, 3D feature films, video games), to industrial design (modelling and visualizing prototypes) and virtual reality (flight simulator, interactive walk-trough). This course introduces the domain by presenting the bases for the creation of 3D models, their animation, and the rendering of the corresponding 3D scene. Student will be invited to practice through programming exercises in OpenGL.

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Content

This course include practical sessions.

Visit the full page about this course

Visit the full page about this course

This course provides an introduction to the basic concepts of object-oriented programming in C++: classes and encapsulation, operator overloading, generic classes (templates), STL (Standard Template Library), inheritance and derived classes, polymorphism and virtual functions.
Lab sessions illustrate these concepts, and applications for applied mathematics. 

Visit the full page about this course

Contents:

• Types of equations, conservation laws
• Finite differences methods
• Laplace equation
• Parabolic equations (diffusion)
• Hyperbolic equations (propagation)
• Non linear hyperbolic equations

This course include practical sessions.

This is a two parts course, this part is the course mutualized with Ensimag 2A 4MMMEDPS.

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Contents:

• Image definition
• Fourier transform, FFT, applications
• Image digitalisation, sampling
• Image processing: convolution, filtering. Applications
• Image decomposition, multiresolution. Application to compression

This course includes practical sessions.

Visit the full page about this course

This course is an introduction to the differential geometry of curves and surfaces with a particular focus on spline curves and surfaces that are routinely used in geometrical design softwares.

Content

This course includes practical sessions.

Visit the full page about this course

Connaitre les notions élémentaires et indispensables de probabilités. Maitriser les calculs de lois, d'espérance; connaitre les fonctions de simulation de logiciels spécialisés; savoir utiliser ces fonctions pour illustrer les résultats
de probabilité. Chaque notion sera abordée en cours et illustrée numériquement en séance
avec les étudiants.

Connaitre les principaux estimateurs, leurs propriétés.
Être capable de programmer les différents estimateurs dans des
situations simples; savoir illustrer par simulation les propriétés des estimateurs (biais, erreur quadratique).
Chaque notion sera abordée en cours et illustrée numériquement en séance avec les étudiants.

Contenu : Lois fondamentales discrètes et continues; Variables indépendantes, lois conditionnelles.
Espérance, espérance conditionnelle. Convergence de variables aléatoires. Théorème
centrale limite, loi des grands nombres. Vecteur gaussien, théorème Cochran. Notion
de chaine de Markov illustrée dans des TP.

Estimateur, estimation. Biais, erreur quadratique, convergence. Intervalle de
confiance. Estimateur des moindres carrés. Estimateur des moments. Estimateur du maximum
de vraisemblance. Equations estimantes.

Visit the full page about this course

The syllabus of the English course in M2 aims at enabling students to validate 3 competences that will be essential for their working life or for their doctoral studies in the future, at the B2 level:

1°) CAN give a clear presentation on a familiar topic, and CAN answer predictable or factual questions

2°) CAN find relevant information and essential points in written texts

3°) CAN make simple notes that are of reasonable use for essay or revision purposes, capturing most important points." (CEF, appendix D).

Reading comprehension can be validated in M1.

The course contents are linked to the students' fields of studies.

Visit the full page about this course

Contents:

• Introduction to database
• Introduction to big data
• Introduction to high performance computing (HPC)
• Numerical solvers for HPC

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

January science and/or industrial project.

Visit the full page about this course

Industrial and/or research internship.

Visit the full page about this course

This program combines case studies coming from real life problems or models and lectures providing the mathematical and numerical backgrounds.

Contents:

• Introduction, classification, examples.
• Theoretical results: convexity and compacity, optimality conditions, KT theorem
• Algorithmic for unconstrained optimisation (descent, line search, (quasi) Newton)
• Algorithms for non differentiable problems
• Algorithms for constrained optimisation: penalisatio, SQP methods
• Applications

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Operational Research proposes scientific methods
to help make better decisions. The idea is to develop and use mathematical and computer
tools to master complex problems. Practical applications are historically in
the direction and management of large systems of people, machines, and
materials in industry, service, humanitarian aid, environment ...

In this course, we will focus on especially on problems with a combinatorial
structure: the number of possible solutions is finite but too large to be
enumerated. The study of these problems involves a phase of modelling practical
problems and then algorithmic resolution.

At the end of this course, students will be able to propose a model and will be
able to implement practical solutions (dedicated or industrial tools) to deal
with a problem of decision or optimization.

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

To acquire the main theoretical and practical notions of modern cryptography: from notions in algorithmic complexity and information theory, to a general overview on the main algorithms and protocols in symmetric and asymmetric cryptography.

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Computer Graphics covers the set of techniques enabling the synthesis of animated virtual worlds. The applications range from entertainment (special effects, 3D feature films, video games), to industrial design (modelling and visualizing prototypes) and virtual reality (flight simulator, interactive walk-trough). This course introduces the domain by presenting the bases for the creation of 3D models, their animation, and the rendering of the corresponding 3D scene. Student will be invited to practice through programming exercises in OpenGL.

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Content

This course include practical sessions.

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

Visit the full page about this course

The course is structured in two parts, treated respectively and independently by Sylvain Meignen and Kevin Polisano. The first part is devoted to differential calculus and its applications in image restoration and edge detection. The second part is dedicated to the construction and practical use of the wavelet transform. Wavelets are basis functions widely used in a large variety of fields: signal and image processing, data compression, smoothing/denoising data, numerical schemes for partial differential equations, scientific visualization, etc. Connections between the two parts will be made on the aspects of denoising, edge detection and graph analysis.

Course outline

Part I: Differential Calculus

1. Differentiability on normed vector spaces
2. Image restoration
3. Edge detection

Part II: Wavelets and Applications

1. From Fourier to the 1D Continuous Wavelet Transform
2. Wavelet zoom, a local characterization of functions
3. The 2D Continuous Wavelet Transform
4. The 1D and 2D Discrete Wavelet Transform
5. Linear and nonlinear approximations in wavelet bases
6. The graph Fourier and wavelets transforms

Visit the full page about this course

In a very broad acceptation, shape and topology optimization is about finding the best domain (which may represent, depending on applications, a mechanical structure, a fluid channel,…) with respect to a given performance criterion (e.g. robustness, weight, etc.), under some constraints (e.g. of a geometric nature). Fostered by its impressive technological and industrial achievements, this discipline has aroused a growing enthusiasm among mathematicians, physicists and engineers since the seventies. Nowadays, problems pertaining to fields so diverse as mechanical engineering, fluid mechanics or biology, to name a few, are currently tackled with optimal design techniques, and constantly raise new, challenging issues.

Visit the full page about this course

The subject of this half-semester course are more advanced methods in convex optimization. It consists of 6 lectures, 2 x 1,5 hours each, and can be seen as continuation of the course “Non-smooth methods in convex optimization”.

This course deals with:

Evaluation : A two-hours written exam (E1) in December. For those who do not pass there will be another two-hours exam (E2) in session 2 in spring.

Visit the full page about this course

This interdisciplinary MSc course is designed for applicants with a biomedical, computational or mathematical background. It provides students with the necessary skills to produce effective research in bioinformatics and computational biology.

Visit the full page about this course

The first part of the lecture introduce to mathematical modeling of fluid mechanics and the numerical resolution of the associated equations. Equations are classified by three main families of models:

1. environmental problems: yield stress fluids (Bingham type) for granular matter, e.g. snow avalanches, mud or ice flows, erosion, landslides and volcanic lavas.

2. industrial problems: viscoelastic fluids (Oldroyd type) for plastic material processes, and metallic alloy.

3. biological problems: elastoviscoplastic fluids, for blood flows, liquid foam flows, and for food processing (mayonnaise, ketchup, etc).

Equations and models are presented in a continuum setting, and then approximated in time and space. Then, the efficient numerical resolution is addressed with some examples of practical applications.

The second part of the lecture propose a deeper analysis of granular models. The mathematical study of these complex matter is an important numerical and physical challenge. We will show how it requires a general view related to nonlinear PDEs. The objective of this course will be two-fold:

• Show how the compressibility and the viscoplasticity of the phenomenon can play an important role
• Discuss congestion phenomena in granular media (maximum packing) that can be compared mathematically to floating structure phenomena in the presence of a free boundary.

Students who complete the course will have demonstrated the ability to do the following:

• formulate and solve a large number of nonlinear physical and mechanical problems.
• demonstrate a familiarity with fluid mechanics and complex materials
• synthesize and implement efficient algorithms for various applications of industrial type.

The main idea of this lecture is to motivate by examples interdisciplinary collaborations needed to deal with complex situations.

Visit the full page about this course

In this course, we will introduce parallel programming paradigms to the students in the context of applied mathematics. The students will learn to identify the parallel pattern in numerical algorithm. The key components that the course will focus on are : efficiency, scalability, parallel pattern, comparison of parallel algorithms, operational intensity and emerging programming paradigm. Trough different lab assignments, the students will apply the concepts of efficient parallel programming using Graphic Processing Unit. In the final project, the students will have the possibility to parallelize one of their own numerical application developed in a previous course.

Visit the full page about this course

The aim of this course is to study various useful applications, libraries and methods for software engineering related to applied mathematics. For example :

•        C++ project management (git and/or svn)
•        Development and profiling
•        Boost library
•        Linear algebra (Eigen)
•        Prototyping and interfacing using Python
•        Post processing and visualization tools (VTK, Paraview, GMSH)

This course deals with :

Topic 1: Software Engineering
Topic 2: Programming

Evaluation :

Practial sessions reports and oral presentation at the end of the course

Visit the full page about this course

In the current context of energy transition and fight against global warming, a precise knowledge of the crust, down to several km depth, has become a critical issue. The crust is the place where are to be found ore resources needed to build electric batteries (rare earth elements) as well as concrete resources for offshore and onshore wind turbines foundations. The crust is also the only place presenting sufficient volumes to store CO2 and H2 in a flexible way. CO2 storage will be a crucial component among industrial solutions to fight against global warming and reach neutral carbon emissions in the next decades.

To these ends, high resolution quantitative estimates of the mechanical parameters of the crust is essential. To perform such estimation, one has to rely on the interpretation of the mechanical waves which travel in the crust. The inference of the mechanical properties of the subsurface from local recording of the mechanical waves at the surface is a mathematical inverse problem. The aim of this course is to provide the mathematical background and the required theoretical tools to introduce high resolution seismic imaging methods to the students, complemented with practical numerical work on schematic examples.

The first main part of the course will be devoted to the theoretical and practical aspects of wave propagation in heterogeneous media. Beginning by some general consideration on hyperbolic partial differential equations, we will see how the elastodynamics equations, representing the propagation of mechanical waves in the subsurface, belong to this category of equations. We will show in particular an energy conservation result based on the symmetry of the underlying hyperbolic system. We will then discuss how to design absorbing boundary conditions for wave propagation problems, to mimic media of infinite extension. This will lead us to the question of numerical approximation to the solution of wave equations in heterogeneous media. We will discuss in details finite-difference schemes, and practical work will be dedicated to the implementation of a finite-difference scheme for the 1D and 2D acoustic equations, and potentially 2D elastic equations.

The second main part of the course will be devoted to the theoretical and practical aspects of seismic imaging using full waveform inversion. We will show how this method is formulated as a nonlinear inverse problem, controlled by partial differential equations representing wave propagation in heterogeneous media. We will discuss how this problem can be solved by local optimization strategies, and review such strategies, from 1st order gradient method to more evolved 2nd order Newton or quasi-Newton methods. The computation of the gradient of the misfit function through the adjoint state method, following optimal control theory, will be extensively presented, as well as its physical interpretation. This theoretical work will be supported by numerical experiments based on the finite-difference wave propagation code developed in the first part of the course. We will then discuss how full waveform inversion is applied in practice, supported by various field data applications examples. This will lead us to discuss current limitations of the method related to its ill-posedness and the lack of regularity of the solution, and give an overview of methodological work currently performed to mitigate these limitations.

Course outline

2 introductory session

• main introduction on seismic imaging (to do what?)
• main concepts related to general inverse problems

5 modeling sessions

• theoretical considerations on hyperbolic systems
• how to derive the elastodynamics equations from Newton and Hooke’s law
• elastodynamics equations = symmetrizable hyperbolic system, energy conservation
• absorbing boundary conditions - numerical approximation to the solution of wave propagation in heterogeneous media (finite-difference, finite element) - practical work : implement 1D and 2D acoustic, + 2D elastic if time allows

5 inverse problem sessions

• imaging the crust= nonlinear inverse problem controlled by an hyperbolic PDE
• local optimization method
• gradient computation through the adjoint state strategy
• physical interpretation of the gradient and Hessian operators - implementation of the gradient computation based on the modeling code designed in the first part
• full waveform inversion in practice: hierarchical schemes
• review of applications - review of current methodological developments

Visit the full page about this course

Numerical simulation is ubiquitous in today’s world. Initially confined to well-mastered physical problems, it has spread to all fields (oceanography, biology, ecology, etc.), the aim being to make forecasts of the systems under study. This has been possible thanks to the combination of numerical models and access to a considerable amount of data. However, there are many sources of uncertainty in these modelling systems. They can come from poorly known processes, approximations in the model equations and/or in their discretization, partial and uncertain data, … The objective of this course is to explore in depth the mathematical methods that have allowed these two worlds to meet. Firstly, we will focus on sensitivity analysis approaches that allow us to study the behavior of the system and its response to perturbations. In particular, this permits to study the way in which uncertainties are propagated. Next, we will look at data assimilation methods that aim at reducing said uncertainties by combining numerical models and observation data. Finally, the notions of model reduction will be discussed, which allow the implementation of the previous methods on high dimensional problems.

This course is intended for DS and MSCI students and will start with a differentiated refresher course on the necessary basic mathematical notions.

Course outline

• General introduction and reminder of the basic concepts

• Sensitivity analysis

  • Local sensitivity analysis

  • Global sensitivity analysis

• Data assimilation

  • Variational methods

  • Stochastic methods

• Model reduction

  • Gaussian processes

  • Polynomial Chaos

Visit the full page about this course

This lecture proposes modelling problems. The problems can be industrial or academic. Students are faced to an industrial problem or an academic problem (research oriented). They are in charge of this project. An teacher/tutor may guide them to find solutions to the problem. For industrial project, they have to understand the user needs, to analyze and model the problem, to derive specifications, to implement a solution and to develop the communication and the presentation of the proposed solution. More academic projects are linked to the courses. They are constructed such that the students can go deeper into a subject.

This lecture introduces basic communication methodes in industry. This part is in french and optional.

Rules: the students have to choose TWO subjects (either academic or industrial). They work in small groups on both projects with tutor (analysis of the problem, bibliography, construction of a solution, numerical simulations, etc.). At the end, they defend their results in front of a jury and provide a short report.

Visit the full page about this course

The quantum formalism developed a century ago provides a very precise description of nature at small scales which entails several counter intuitive aspects: superposition of states, entanglement, intrinsic randomness of measurement process, to list a few. However, from a mathematical point of view, quantum mechanics does have a definite formulation. This allows to investigate these intriguing features rigorously and to explore these quantum traits in information theory and algorithmics in particular, as well as the challenges they present.

The goal of these lectures is to provide a mathematical description of the quantum formalism in finite dimension and to introduce the mathematical concepts and tools required for the analysis of such quantum systems and their dynamics. On the one hand, we will study the key aspects of quantum information theory. On the other hand, we will describe certain properties of quantum dynamics that need to be taken into account in the implementation of quantum algorithms and that will be applied to emblematic systems. The interaction with an external classical electromagnetic field will also be considered both from a theoretical and a numerical point of view.

Visit the full page about this course

The goal of this course is to present a wide range of recent numerical methods and algorithms that find applications in various fields. More precisely, the course will focus on optimal transport algorithms, proximal methods and level set methods – the leading application of these being image analysis.

Optimal transport is an important field of mathematics that was originally introduced in the 1700’s by the French mathematician and engineer Gaspard Monge to answer the following concrete question: what is the cheapest way of sending a pile of sand into a hole, knowing the cost of transportation of each sand grain of the pile to a possible target location? This problem gave the birth of the theory of optimal transport. This theory has connections with PDEs, geometry and probability and has been used in many fields such as computer vision, economy, non-imaging optics… In the last 15 years, this problem has been extensively studied from a computational point of view and different efficient algorithms have been proposed.

In this course, we will provide an introduction to the optimal transport theory and the analysis of several algorithms dedicated to the discrete case (i.e. when the source and target measures are discrete), such as Auction’s algorithm and Sinkorn algorithm. We will also study the semi-discrete setting that corresponds to transporting a continuous measure to a discrete one and we will analyze Oliker-Prüssner algorithm as well as a Newton algorithm.

We will also consider some numerical methods which have wide applications in several modeling fields as the Level Set method to capture interfaces, primal dual methods, with main application in this course to image analysis : active contours, deblurring, demonising, inpainting and interpolation, the latter issue being dealt with by a so-called dynamic formulation of optimal transport.

Visit the full page about this course

This course is related to mathematical and statistical methods which are very used in supervised learning.

It contains two parts.

In the first part, we will focus on parametric modeling. Starting with the classical linear regression, we will describe several families of estimators that work when considering high-dimensional data, where the classical least square estimator does not work. Model selection and model assessment will particularly be described.

In the second part, we shall focus on nonparametric methods. We will present several tools and ingredients to predict the future value of a variable. We shall focus on methods for non parametric regression from independent to correlated training dataset. We shall also study some methods to avoid the overfitting in supervised learning.

This course will be followed by practical sessions with the R software.

Visit the full page about this course

Modelling extreme temperatures, extreme river flows, earthquakes intensities, neuronal activity, map diseases, lightning strikes, forest fires, for example is a risk modelling and assessment task, which is tackled in statistics using point processes and extreme value theory.

On the one hand, point processes are a class of stochastic processes modelling random events in interaction. By event we can think of the time a neuron activates, an earthquake occurs, the time a tweet has been retweeted, etc or the location of a tree in a forest, the impact of a lightning strike, etc. The first two parts provide an introduction to stochastic models and statistical inference which could cover such applications. Main characteristics of such processes, standard models (properties, simulation) and statistical procedures to infer them will be presented.

On the other hand, taking into account extreme events such as heavy rainfalls, floods, extreme temperatures is often crucial in the statistical approach to risk modeling. In this context, the behavior of the distribution tail is then more important than the shape of the central part of the distribution. Extreme-value theory offers a wide range of tools for modeling and estimating the probability of extreme events.

Visit the full page about this course

This is the master thesis project.

Visit the full page about this course

The course is split into two parts. During the first part, a wide range of machine learning algorithms will be discussed. The second part will focus on deep learning, and presentations more applied to the three data modalities and their combinations. The following is a non-exhaustive list of topics discussed:

• Computing dot products in high dimension & Page Rank
• Matrix completion/factorization (Stochastic Gradient Descent, SVD)
• Monte-carlo, MCMC methods: Metropolis-Hastings and Gibbs Sampling
• Unsupervised classification: Partitionning, Hierarchical, Kernel and Spectral clustering
• Alignment and matching algorithms (local/global, pairwise/multiple), dynamic programming, Hungarian algorithm,…
• Introduction to Deep Learning concepts, including CNN, RNN, Metric learning
• Attention models: Self-attention, Transformers
• Auditory data: Representation, sound source localisation and separation.
• Natural language data: Representation, Seq2Seq, Word2Vec, Machine Translation, Pre-training strategies, Benchmarks and evaluation
• Visual data: image and video representation, recap of traditional features, state-of-the-art neural architectures for feature extraction
• Object detection and recognition, action recognition.
• Multimodal learning: audio-visual data representation, multimedia retrieval.
• Generative Adversarial Networks: Image-image translation, conditional generation

Visit the full page about this course

In a very broad acceptation, shape and topology optimization is about finding the best domain (which may represent, depending on applications, a mechanical structure, a fluid channel,…) with respect to a given performance criterion (e.g. robustness, weight, etc.), under some constraints (e.g. of a geometric nature). Fostered by its impressive technological and industrial achievements, this discipline has aroused a growing enthusiasm among mathematicians, physicists and engineers since the seventies. Nowadays, problems pertaining to fields so diverse as mechanical engineering, fluid mechanics or biology, to name a few, are currently tackled with optimal design techniques, and constantly raise new, challenging issues.

Visit the full page about this course

This interdisciplinary MSc course is designed for applicants with a biomedical, computational or mathematical background. It provides students with the necessary skills to produce effective research in bioinformatics and computational biology.

Visit the full page about this course

This course contains two parts.

Part I concerns Data challenge.

This part consists in a real problem that is given to the students for which data are readily available. The goal is to have teams of five to six students compete in solving (at least partially) the problem.

The work is spread over the Autumn semester and consists of: building a prediction model or a methodology to solve the problem based on a set of training data, blind evaluation of the model or methodology on a test bench (unseen data, withheld from the students), using an appropriate performance measure.

At the end, the teams will present their solution path in a formal presentation and a short report.

Part II concerns Data Science seminars.

This is a cycle of seminars or presentations with a common factor that is the project of the data challenge. A first seminar will settle the context and the problem for that year’s data challenge.

The other seminars will propose different industrial or academic approaches and problems that are (loosely) related to the objective of the data challenge. Presentations have a time slot of one hour and students will have to read up front some ressources to orient their questions about the subject after the seminar.

Visit the full page about this course

The course is structured in two parts, treated respectively and independently by Sylvain Meignen and Kevin Polisano. The first part is devoted to differential calculus and its applications in image restoration and edge detection. The second part is dedicated to the construction and practical use of the wavelet transform. Wavelets are basis functions widely used in a large variety of fields: signal and image processing, data compression, smoothing/denoising data, numerical schemes for partial differential equations, scientific visualization, etc. Connections between the two parts will be made on the aspects of denoising, edge detection and graph analysis.

Course outline

Part I: Differential Calculus

1. Differentiability on normed vector spaces
2. Image restoration
3. Edge detection

Part II: Wavelets and Applications

1. From Fourier to the 1D Continuous Wavelet Transform
2. Wavelet zoom, a local characterization of functions
3. The 2D Continuous Wavelet Transform
4. The 1D and 2D Discrete Wavelet Transform
5. Linear and nonlinear approximations in wavelet bases
6. The graph Fourier and wavelets transforms

Visit the full page about this course

The subject of this half-semester course are more advanced methods in convex optimization. It consists of 6 lectures, 2 x 1,5 hours each, and can be seen as continuation of the course “Non-smooth methods in convex optimization”.

This course deals with:

Evaluation : A two-hours written exam (E1) in December. For those who do not pass there will be another two-hours exam (E2) in session 2 in spring.

Visit the full page about this course

Statistical learning is about the construction and study of systems that can automatically learn from data. With the emergence of massive datasets commonly encountered today, the need for powerful machine learning is of acute importance. Examples of successful applications include effective web search, anti-spam software, computer vision, robotics, practical speech recognition, and a deeper understanding of the human genome. This course gives an introduction to this exciting field. In the first part, we will introduce basic techniques such as logistic regression, multilayer perceptrons, nearest neighbor approaches, both from a theoretical and methodological point of views. In the second part, we will focus on more advanced techniques such as kernel methods, which is a versatile tool to represent data, in combination with (un)supervised learning techniques that are agnostic to the type of data that is learned from. The learning techniques that will be covered include regression, classification, clustering and dimension reduction. We will cover both the theoretical underpinnings of kernels, as well as a series of kernels that are important in practical applications. Finally we will touch upon topics of active research, such as large-scale kernel methods and the use of kernel methods to develop theoretical foundations of deep learning models.

Visit the full page about this course

Numerical simulation is ubiquitous in today’s world. Initially confined to well-mastered physical problems, it has spread to all fields (oceanography, biology, ecology, etc.), the aim being to make forecasts of the systems under study. This has been possible thanks to the combination of numerical models and access to a considerable amount of data. However, there are many sources of uncertainty in these modelling systems. They can come from poorly known processes, approximations in the model equations and/or in their discretization, partial and uncertain data, … The objective of this course is to explore in depth the mathematical methods that have allowed these two worlds to meet. Firstly, we will focus on sensitivity analysis approaches that allow us to study the behavior of the system and its response to perturbations. In particular, this permits to study the way in which uncertainties are propagated. Next, we will look at data assimilation methods that aim at reducing said uncertainties by combining numerical models and observation data. Finally, the notions of model reduction will be discussed, which allow the implementation of the previous methods on high dimensional problems.

This course is intended for DS and MSCI students and will start with a differentiated refresher course on the necessary basic mathematical notions.

Course outline

• General introduction and reminder of the basic concepts

• Sensitivity analysis

  • Local sensitivity analysis

  • Global sensitivity analysis

• Data assimilation

  • Variational methods

  • Stochastic methods

• Model reduction

  • Gaussian processes

  • Polynomial Chaos

Visit the full page about this course

In this course, we will introduce parallel programming paradigms to the students in the context of applied mathematics. The students will learn to identify the parallel pattern in numerical algorithm. The key components that the course will focus on are : efficiency, scalability, parallel pattern, comparison of parallel algorithms, operational intensity and emerging programming paradigm. Trough different lab assignments, the students will apply the concepts of efficient parallel programming using Graphic Processing Unit. In the final project, the students will have the possibility to parallelize one of their own numerical application developed in a previous course.

Visit the full page about this course

Causality is at the core of our vision of the world and of the way we reason. It has long been recognized as an important concept and was already mentioned in the ancient Hindu scriptures: “Cause is the effect concealed, effect is the cause revealed”. Even Democritus famously proclaimed that he would rather discover a causal relation than be the king of presumably the wealthiest empire of his time. Nowadays, causality is seen as an ideal way to explain observed phenomena and to provide tools to reason on possible outcomes of interventions and what-if experiments, which are central to counterfactual reasoning, as ‘‘what if this patient had been given this particular treatment?’’

Visit the full page about this course

Machine Learning is one of the key areas of Artificial Intelligence and it concerns the study and the development of quantitative models that enables a computer to perform tasks without being explicitly programmed to do them. Learning in this context is hence to recognize complex forms and to make intelligent decisions. Given all existing entries, the difficulty of this task lies in the fact that all possible decisions is usually very complex to enumerate. To get around that, machine learning algorithms are designed in order to gain knowledge on the problem to be addressed based on a limited set of observed data extracted from this problem. To illustrate this principle, consider the supervised learning task, where the prediction function, which infers a predicted output for a given input, is learned over a finite set of labeled training examples, where each instance of this set is a pair constituted of a vector characterizing an observation in a given vector space, and an associated desired response for that instance (also called desired output). After the training step, the function returned by the algorithm is sought to give predictions on new examples, which have not been used in the learning process, with the lowest probability of error. The underlying assumption in this case is that the examples are, in general, representative of the prediction problem on which the function will be applied. We expect that the learning algorithm produces a function that will have a good generalization performance and not the one that is able to perfectly reproduce the outputs associated to the training examples. Guarantees of learnability of this process were studied in the theory of machine learning largely initiated by Vladimir Vapnik. These guarantees are dependent on the size of the training set and the complexity of the class of functions where the algorithm searches for the prediction function. Emerging technologies, particularly those related to the development of Internet, reshaped the domain of machine learning with new learning frameworks that have been studied to better tackle the related problems. One of these frameworks concerns the problem of learning with partially labeled data, or semi-supervised learning, which development is motivated by the effort that has to be made to construct labeled training sets for some problems, while large amount of unlabeled data can be gathered easily for these problems. The inherent assumption, in this case, is that unlabeled data contain relevant information about the task that has to be solved, and that it is a natural idea to try to extract this information so as to provide the learning algorithm more evidence. From these facts were born a number of works that intended to use a small amount of labeled data simultaneously with a large amount of unlabeled data to learn a prediction function.

The intent of this course is to propose a broad introduction to the field of Machine Learning, including discussions of each of the major frameworks, supervised, unsupervised, semi-supervised and reinforcement learning.

Visit the full page about this course

This lecture proposes modelling problems. The problems can be industrial or academic. Students are faced to an industrial problem or an academic problem (research oriented). They are in charge of this project. An teacher/tutor may guide them to find solutions to the problem. For industrial project, they have to understand the user needs, to analyze and model the problem, to derive specifications, to implement a solution and to develop the communication and the presentation of the proposed solution. More academic projects are linked to the courses. They are constructed such that the students can go deeper into a subject.

This lecture introduces basic communication methodes in industry. This part is in french and optional.

Rules: the students have to choose TWO subjects (either academic or industrial). They work in small groups on both projects with tutor (analysis of the problem, bibliography, construction of a solution, numerical simulations, etc.). At the end, they defend their results in front of a jury and provide a short report.

Visit the full page about this course

The goal of this course is to present a wide range of recent numerical methods and algorithms that find applications in various fields. More precisely, the course will focus on optimal transport algorithms, proximal methods and level set methods – the leading application of these being image analysis.

Optimal transport is an important field of mathematics that was originally introduced in the 1700’s by the French mathematician and engineer Gaspard Monge to answer the following concrete question: what is the cheapest way of sending a pile of sand into a hole, knowing the cost of transportation of each sand grain of the pile to a possible target location? This problem gave the birth of the theory of optimal transport. This theory has connections with PDEs, geometry and probability and has been used in many fields such as computer vision, economy, non-imaging optics… In the last 15 years, this problem has been extensively studied from a computational point of view and different efficient algorithms have been proposed.

In this course, we will provide an introduction to the optimal transport theory and the analysis of several algorithms dedicated to the discrete case (i.e. when the source and target measures are discrete), such as Auction’s algorithm and Sinkorn algorithm. We will also study the semi-discrete setting that corresponds to transporting a continuous measure to a discrete one and we will analyze Oliker-Prüssner algorithm as well as a Newton algorithm.

We will also consider some numerical methods which have wide applications in several modeling fields as the Level Set method to capture interfaces, primal dual methods, with main application in this course to image analysis : active contours, deblurring, demonising, inpainting and interpolation, the latter issue being dealt with by a so-called dynamic formulation of optimal transport.

Visit the full page about this course

The automatic processing of languages, whether written or spoken, has always been an essential part of artificial intelligence. This domain has encouraged the emergence of new uses thanks to the arrival in the industrial field of many technologies from research (spell-checkers, speech synthesis, speech recognition, machine translation, …). In this course, we present the most recent advances and challenges for research. We will discuss discourse analysis whether written or spoken, text clarification, automatic speech transcription and automatic translation, in particular recent advances with neural models.

Information access and retrieval is now ubiquitous in everyday life through search engines, recommendation systems, or technological and commercial surveillance, in many application domains either general or specific like health for instance. In this course, we will cover Information retrieval basics, information retrieval evaluation, models for information retrieval, medical information retrieval, and deep learning for multimedia indexing and retrieval

Visit the full page about this course

This course is related to mathematical and statistical methods which are very used in supervised learning.

It contains two parts.

In the first part, we will focus on parametric modeling. Starting with the classical linear regression, we will describe several families of estimators that work when considering high-dimensional data, where the classical least square estimator does not work. Model selection and model assessment will particularly be described.

In the second part, we shall focus on nonparametric methods. We will present several tools and ingredients to predict the future value of a variable. We shall focus on methods for non parametric regression from independent to correlated training dataset. We shall also study some methods to avoid the overfitting in supervised learning.

This course will be followed by practical sessions with the R software.

Visit the full page about this course

The aim of this course is to study various useful applications, libraries and methods for software engineering related to applied mathematics. For example :

•        C++ project management (git and/or svn)
•        Development and profiling
•        Boost library
•        Linear algebra (Eigen)
•        Prototyping and interfacing using Python
•        Post processing and visualization tools (VTK, Paraview, GMSH)

This course deals with :

Topic 1: Software Engineering
Topic 2: Programming

Evaluation :

Practial sessions reports and oral presentation at the end of the course

Visit the full page about this course

Modelling extreme temperatures, extreme river flows, earthquakes intensities, neuronal activity, map diseases, lightning strikes, forest fires, for example is a risk modelling and assessment task, which is tackled in statistics using point processes and extreme value theory.

On the one hand, point processes are a class of stochastic processes modelling random events in interaction. By event we can think of the time a neuron activates, an earthquake occurs, the time a tweet has been retweeted, etc or the location of a tree in a forest, the impact of a lightning strike, etc. The first two parts provide an introduction to stochastic models and statistical inference which could cover such applications. Main characteristics of such processes, standard models (properties, simulation) and statistical procedures to infer them will be presented.

On the other hand, taking into account extreme events such as heavy rainfalls, floods, extreme temperatures is often crucial in the statistical approach to risk modeling. In this context, the behavior of the distribution tail is then more important than the shape of the central part of the distribution. Extreme-value theory offers a wide range of tools for modeling and estimating the probability of extreme events.

Visit the full page about this course

This is the master thesis project.

Visit the full page about this course

Visit the full page about this course

The aim of this course is to study various useful applications, libraries and methods for software engineering related to applied mathematics. For example :

•        C++ project management (git and/or svn)
•        Development and profiling
•        Boost library
•        Linear algebra (Eigen)
•        Prototyping and interfacing using Python
•        Post processing and visualization tools (VTK, Paraview, GMSH)

This course deals with :

Topic 1: Software Engineering
Topic 2: Programming

Evaluation :

Practial sessions reports and oral presentation at the end of the course

Visit the full page about this course

This lecture proposes modelling problems. The problems can be industrial or academic. Students are faced to an industrial problem or an academic problem (research oriented). They are in charge of this project. An teacher/tutor may guide them to find solutions to the problem. For industrial project, they have to understand the user needs, to analyze and model the problem, to derive specifications, to implement a solution and to develop the communication and the presentation of the proposed solution. More academic projects are linked to the courses. They are constructed such that the students can go deeper into a subject.

This lecture introduces basic communication methodes in industry. This part is in french and optional.

Rules: the students have to choose TWO subjects (either academic or industrial). They work in small groups on both projects with tutor (analysis of the problem, bibliography, construction of a solution, numerical simulations, etc.). At the end, they defend their results in front of a jury and provide a short report.

Visit the full page about this course

In the current context of energy transition and fight against global warming, a precise knowledge of the crust, down to several km depth, has become a critical issue. The crust is the place where are to be found ore resources needed to build electric batteries (rare earth elements) as well as concrete resources for offshore and onshore wind turbines foundations. The crust is also the only place presenting sufficient volumes to store CO2 and H2 in a flexible way. CO2 storage will be a crucial component among industrial solutions to fight against global warming and reach neutral carbon emissions in the next decades.

To these ends, high resolution quantitative estimates of the mechanical parameters of the crust is essential. To perform such estimation, one has to rely on the interpretation of the mechanical waves which travel in the crust. The inference of the mechanical properties of the subsurface from local recording of the mechanical waves at the surface is a mathematical inverse problem. The aim of this course is to provide the mathematical background and the required theoretical tools to introduce high resolution seismic imaging methods to the students, complemented with practical numerical work on schematic examples.

The first main part of the course will be devoted to the theoretical and practical aspects of wave propagation in heterogeneous media. Beginning by some general consideration on hyperbolic partial differential equations, we will see how the elastodynamics equations, representing the propagation of mechanical waves in the subsurface, belong to this category of equations. We will show in particular an energy conservation result based on the symmetry of the underlying hyperbolic system. We will then discuss how to design absorbing boundary conditions for wave propagation problems, to mimic media of infinite extension. This will lead us to the question of numerical approximation to the solution of wave equations in heterogeneous media. We will discuss in details finite-difference schemes, and practical work will be dedicated to the implementation of a finite-difference scheme for the 1D and 2D acoustic equations, and potentially 2D elastic equations.

The second main part of the course will be devoted to the theoretical and practical aspects of seismic imaging using full waveform inversion. We will show how this method is formulated as a nonlinear inverse problem, controlled by partial differential equations representing wave propagation in heterogeneous media. We will discuss how this problem can be solved by local optimization strategies, and review such strategies, from 1st order gradient method to more evolved 2nd order Newton or quasi-Newton methods. The computation of the gradient of the misfit function through the adjoint state method, following optimal control theory, will be extensively presented, as well as its physical interpretation. This theoretical work will be supported by numerical experiments based on the finite-difference wave propagation code developed in the first part of the course. We will then discuss how full waveform inversion is applied in practice, supported by various field data applications examples. This will lead us to discuss current limitations of the method related to its ill-posedness and the lack of regularity of the solution, and give an overview of methodological work currently performed to mitigate these limitations.

Course outline

2 introductory session

• main introduction on seismic imaging (to do what?)
• main concepts related to general inverse problems

5 modeling sessions

• theoretical considerations on hyperbolic systems
• how to derive the elastodynamics equations from Newton and Hooke’s law
• elastodynamics equations = symmetrizable hyperbolic system, energy conservation
• absorbing boundary conditions - numerical approximation to the solution of wave propagation in heterogeneous media (finite-difference, finite element) - practical work : implement 1D and 2D acoustic, + 2D elastic if time allows

5 inverse problem sessions

• imaging the crust= nonlinear inverse problem controlled by an hyperbolic PDE
• local optimization method
• gradient computation through the adjoint state strategy
• physical interpretation of the gradient and Hessian operators - implementation of the gradient computation based on the modeling code designed in the first part
• full waveform inversion in practice: hierarchical schemes
• review of applications - review of current methodological developments

Visit the full page about this course

In a very broad acceptation, shape and topology optimization is about finding the best domain (which may represent, depending on applications, a mechanical structure, a fluid channel,…) with respect to a given performance criterion (e.g. robustness, weight, etc.), under some constraints (e.g. of a geometric nature). Fostered by its impressive technological and industrial achievements, this discipline has aroused a growing enthusiasm among mathematicians, physicists and engineers since the seventies. Nowadays, problems pertaining to fields so diverse as mechanical engineering, fluid mechanics or biology, to name a few, are currently tackled with optimal design techniques, and constantly raise new, challenging issues.

Visit the full page about this course

Refresh courses

Visit the full page about this course

In this course, we will introduce parallel programming paradigms to the students in the context of applied mathematics. The students will learn to identify the parallel pattern in numerical algorithm. The key components that the course will focus on are : efficiency, scalability, parallel pattern, comparison of parallel algorithms, operational intensity and emerging programming paradigm. Trough different lab assignments, the students will apply the concepts of efficient parallel programming using Graphic Processing Unit. In the final project, the students will have the possibility to parallelize one of their own numerical application developed in a previous course.

Visit the full page about this course

The course is structured in two parts, treated respectively and independently by Sylvain Meignen and Kevin Polisano. The first part is devoted to differential calculus and its applications in image restoration and edge detection. The second part is dedicated to the construction and practical use of the wavelet transform. Wavelets are basis functions widely used in a large variety of fields: signal and image processing, data compression, smoothing/denoising data, numerical schemes for partial differential equations, scientific visualization, etc. Connections between the two parts will be made on the aspects of denoising, edge detection and graph analysis.

Course outline

Part I: Differential Calculus

1. Differentiability on normed vector spaces
2. Image restoration
3. Edge detection

Part II: Wavelets and Applications

1. From Fourier to the 1D Continuous Wavelet Transform
2. Wavelet zoom, a local characterization of functions
3. The 2D Continuous Wavelet Transform
4. The 1D and 2D Discrete Wavelet Transform
5. Linear and nonlinear approximations in wavelet bases
6. The graph Fourier and wavelets transforms

Visit the full page about this course

The goal of this course is to present a wide range of recent numerical methods and algorithms that find applications in various fields. More precisely, the course will focus on optimal transport algorithms, proximal methods and level set methods – the leading application of these being image analysis.

Optimal transport is an important field of mathematics that was originally introduced in the 1700’s by the French mathematician and engineer Gaspard Monge to answer the following concrete question: what is the cheapest way of sending a pile of sand into a hole, knowing the cost of transportation of each sand grain of the pile to a possible target location? This problem gave the birth of the theory of optimal transport. This theory has connections with PDEs, geometry and probability and has been used in many fields such as computer vision, economy, non-imaging optics… In the last 15 years, this problem has been extensively studied from a computational point of view and different efficient algorithms have been proposed.

In this course, we will provide an introduction to the optimal transport theory and the analysis of several algorithms dedicated to the discrete case (i.e. when the source and target measures are discrete), such as Auction’s algorithm and Sinkorn algorithm. We will also study the semi-discrete setting that corresponds to transporting a continuous measure to a discrete one and we will analyze Oliker-Prüssner algorithm as well as a Newton algorithm.

We will also consider some numerical methods which have wide applications in several modeling fields as the Level Set method to capture interfaces, primal dual methods, with main application in this course to image analysis : active contours, deblurring, demonising, inpainting and interpolation, the latter issue being dealt with by a so-called dynamic formulation of optimal transport.

Visit the full page about this course

The first part of the lecture introduce to mathematical modeling of fluid mechanics and the numerical resolution of the associated equations. Equations are classified by three main families of models:

1. environmental problems: yield stress fluids (Bingham type) for granular matter, e.g. snow avalanches, mud or ice flows, erosion, landslides and volcanic lavas.

2. industrial problems: viscoelastic fluids (Oldroyd type) for plastic material processes, and metallic alloy.

3. biological problems: elastoviscoplastic fluids, for blood flows, liquid foam flows, and for food processing (mayonnaise, ketchup, etc).

Equations and models are presented in a continuum setting, and then approximated in time and space. Then, the efficient numerical resolution is addressed with some examples of practical applications.

The second part of the lecture propose a deeper analysis of granular models. The mathematical study of these complex matter is an important numerical and physical challenge. We will show how it requires a general view related to nonlinear PDEs. The objective of this course will be two-fold:

• Show how the compressibility and the viscoplasticity of the phenomenon can play an important role
• Discuss congestion phenomena in granular media (maximum packing) that can be compared mathematically to floating structure phenomena in the presence of a free boundary.

Students who complete the course will have demonstrated the ability to do the following:

• formulate and solve a large number of nonlinear physical and mechanical problems.
• demonstrate a familiarity with fluid mechanics and complex materials
• synthesize and implement efficient algorithms for various applications of industrial type.

The main idea of this lecture is to motivate by examples interdisciplinary collaborations needed to deal with complex situations.

Visit the full page about this course

Numerical simulation is ubiquitous in today’s world. Initially confined to well-mastered physical problems, it has spread to all fields (oceanography, biology, ecology, etc.), the aim being to make forecasts of the systems under study. This has been possible thanks to the combination of numerical models and access to a considerable amount of data. However, there are many sources of uncertainty in these modelling systems. They can come from poorly known processes, approximations in the model equations and/or in their discretization, partial and uncertain data, … The objective of this course is to explore in depth the mathematical methods that have allowed these two worlds to meet. Firstly, we will focus on sensitivity analysis approaches that allow us to study the behavior of the system and its response to perturbations. In particular, this permits to study the way in which uncertainties are propagated. Next, we will look at data assimilation methods that aim at reducing said uncertainties by combining numerical models and observation data. Finally, the notions of model reduction will be discussed, which allow the implementation of the previous methods on high dimensional problems.

This course is intended for DS and MSCI students and will start with a differentiated refresher course on the necessary basic mathematical notions.

Course outline

• General introduction and reminder of the basic concepts

• Sensitivity analysis

  • Local sensitivity analysis

  • Global sensitivity analysis

• Data assimilation

  • Variational methods

  • Stochastic methods

• Model reduction

  • Gaussian processes

  • Polynomial Chaos

Visit the full page about this course

Modelling extreme temperatures, extreme river flows, earthquakes intensities, neuronal activity, map diseases, lightning strikes, forest fires, for example is a risk modelling and assessment task, which is tackled in statistics using point processes and extreme value theory.

On the one hand, point processes are a class of stochastic processes modelling random events in interaction. By event we can think of the time a neuron activates, an earthquake occurs, the time a tweet has been retweeted, etc or the location of a tree in a forest, the impact of a lightning strike, etc. The first two parts provide an introduction to stochastic models and statistical inference which could cover such applications. Main characteristics of such processes, standard models (properties, simulation) and statistical procedures to infer them will be presented.

On the other hand, taking into account extreme events such as heavy rainfalls, floods, extreme temperatures is often crucial in the statistical approach to risk modeling. In this context, the behavior of the distribution tail is then more important than the shape of the central part of the distribution. Extreme-value theory offers a wide range of tools for modeling and estimating the probability of extreme events.

Visit the full page about this course

The course is split into two parts. During the first part, a wide range of machine learning algorithms will be discussed. The second part will focus on deep learning, and presentations more applied to the three data modalities and their combinations. The following is a non-exhaustive list of topics discussed:

• Computing dot products in high dimension & Page Rank
• Matrix completion/factorization (Stochastic Gradient Descent, SVD)
• Monte-carlo, MCMC methods: Metropolis-Hastings and Gibbs Sampling
• Unsupervised classification: Partitionning, Hierarchical, Kernel and Spectral clustering
• Alignment and matching algorithms (local/global, pairwise/multiple), dynamic programming, Hungarian algorithm,…
• Introduction to Deep Learning concepts, including CNN, RNN, Metric learning
• Attention models: Self-attention, Transformers
• Auditory data: Representation, sound source localisation and separation.
• Natural language data: Representation, Seq2Seq, Word2Vec, Machine Translation, Pre-training strategies, Benchmarks and evaluation
• Visual data: image and video representation, recap of traditional features, state-of-the-art neural architectures for feature extraction
• Object detection and recognition, action recognition.
• Multimodal learning: audio-visual data representation, multimedia retrieval.
• Generative Adversarial Networks: Image-image translation, conditional generation

Visit the full page about this course

The automatic processing of languages, whether written or spoken, has always been an essential part of artificial intelligence. This domain has encouraged the emergence of new uses thanks to the arrival in the industrial field of many technologies from research (spell-checkers, speech synthesis, speech recognition, machine translation, …). In this course, we present the most recent advances and challenges for research. We will discuss discourse analysis whether written or spoken, text clarification, automatic speech transcription and automatic translation, in particular recent advances with neural models.

Information access and retrieval is now ubiquitous in everyday life through search engines, recommendation systems, or technological and commercial surveillance, in many application domains either general or specific like health for instance. In this course, we will cover Information retrieval basics, information retrieval evaluation, models for information retrieval, medical information retrieval, and deep learning for multimedia indexing and retrieval

Visit the full page about this course

Statistical learning is about the construction and study of systems that can automatically learn from data. With the emergence of massive datasets commonly encountered today, the need for powerful machine learning is of acute importance. Examples of successful applications include effective web search, anti-spam software, computer vision, robotics, practical speech recognition, and a deeper understanding of the human genome. This course gives an introduction to this exciting field. In the first part, we will introduce basic techniques such as logistic regression, multilayer perceptrons, nearest neighbor approaches, both from a theoretical and methodological point of views. In the second part, we will focus on more advanced techniques such as kernel methods, which is a versatile tool to represent data, in combination with (un)supervised learning techniques that are agnostic to the type of data that is learned from. The learning techniques that will be covered include regression, classification, clustering and dimension reduction. We will cover both the theoretical underpinnings of kernels, as well as a series of kernels that are important in practical applications. Finally we will touch upon topics of active research, such as large-scale kernel methods and the use of kernel methods to develop theoretical foundations of deep learning models.

Visit the full page about this course

Machine Learning is one of the key areas of Artificial Intelligence and it concerns the study and the development of quantitative models that enables a computer to perform tasks without being explicitly programmed to do them. Learning in this context is hence to recognize complex forms and to make intelligent decisions. Given all existing entries, the difficulty of this task lies in the fact that all possible decisions is usually very complex to enumerate. To get around that, machine learning algorithms are designed in order to gain knowledge on the problem to be addressed based on a limited set of observed data extracted from this problem. To illustrate this principle, consider the supervised learning task, where the prediction function, which infers a predicted output for a given input, is learned over a finite set of labeled training examples, where each instance of this set is a pair constituted of a vector characterizing an observation in a given vector space, and an associated desired response for that instance (also called desired output). After the training step, the function returned by the algorithm is sought to give predictions on new examples, which have not been used in the learning process, with the lowest probability of error. The underlying assumption in this case is that the examples are, in general, representative of the prediction problem on which the function will be applied. We expect that the learning algorithm produces a function that will have a good generalization performance and not the one that is able to perfectly reproduce the outputs associated to the training examples. Guarantees of learnability of this process were studied in the theory of machine learning largely initiated by Vladimir Vapnik. These guarantees are dependent on the size of the training set and the complexity of the class of functions where the algorithm searches for the prediction function. Emerging technologies, particularly those related to the development of Internet, reshaped the domain of machine learning with new learning frameworks that have been studied to better tackle the related problems. One of these frameworks concerns the problem of learning with partially labeled data, or semi-supervised learning, which development is motivated by the effort that has to be made to construct labeled training sets for some problems, while large amount of unlabeled data can be gathered easily for these problems. The inherent assumption, in this case, is that unlabeled data contain relevant information about the task that has to be solved, and that it is a natural idea to try to extract this information so as to provide the learning algorithm more evidence. From these facts were born a number of works that intended to use a small amount of labeled data simultaneously with a large amount of unlabeled data to learn a prediction function.

The intent of this course is to propose a broad introduction to the field of Machine Learning, including discussions of each of the major frameworks, supervised, unsupervised, semi-supervised and reinforcement learning.

Visit the full page about this course

This course is related to mathematical and statistical methods which are very used in supervised learning.

It contains two parts.

In the first part, we will focus on parametric modeling. Starting with the classical linear regression, we will describe several families of estimators that work when considering high-dimensional data, where the classical least square estimator does not work. Model selection and model assessment will particularly be described.

In the second part, we shall focus on nonparametric methods. We will present several tools and ingredients to predict the future value of a variable. We shall focus on methods for non parametric regression from independent to correlated training dataset. We shall also study some methods to avoid the overfitting in supervised learning.

This course will be followed by practical sessions with the R software.

Visit the full page about this course

Causality is at the core of our vision of the world and of the way we reason. It has long been recognized as an important concept and was already mentioned in the ancient Hindu scriptures: “Cause is the effect concealed, effect is the cause revealed”. Even Democritus famously proclaimed that he would rather discover a causal relation than be the king of presumably the wealthiest empire of his time. Nowadays, causality is seen as an ideal way to explain observed phenomena and to provide tools to reason on possible outcomes of interventions and what-if experiments, which are central to counterfactual reasoning, as ‘‘what if this patient had been given this particular treatment?’’

Visit the full page about this course

The subject of this half-semester course are more advanced methods in convex optimization. It consists of 6 lectures, 2 x 1,5 hours each, and can be seen as continuation of the course “Non-smooth methods in convex optimization”.

This course deals with:

Evaluation : A two-hours written exam (E1) in December. For those who do not pass there will be another two-hours exam (E2) in session 2 in spring.

Visit the full page about this course

This course contains two parts.

Part I concerns Data challenge.

This part consists in a real problem that is given to the students for which data are readily available. The goal is to have teams of five to six students compete in solving (at least partially) the problem.

The work is spread over the Autumn semester and consists of: building a prediction model or a methodology to solve the problem based on a set of training data, blind evaluation of the model or methodology on a test bench (unseen data, withheld from the students), using an appropriate performance measure.

At the end, the teams will present their solution path in a formal presentation and a short report.

Part II concerns Data Science seminars.

This is a cycle of seminars or presentations with a common factor that is the project of the data challenge. A first seminar will settle the context and the problem for that year’s data challenge.

The other seminars will propose different industrial or academic approaches and problems that are (loosely) related to the objective of the data challenge. Presentations have a time slot of one hour and students will have to read up front some ressources to orient their questions about the subject after the seminar.

Visit the full page about this course

This interdisciplinary MSc course is designed for applicants with a biomedical, computational or mathematical background. It provides students with the necessary skills to produce effective research in bioinformatics and computational biology.

Visit the full page about this course

The quantum formalism developed a century ago provides a very precise description of nature at small scales which entails several counter intuitive aspects: superposition of states, entanglement, intrinsic randomness of measurement process, to list a few. However, from a mathematical point of view, quantum mechanics does have a definite formulation. This allows to investigate these intriguing features rigorously and to explore these quantum traits in information theory and algorithmics in particular, as well as the challenges they present.

The goal of these lectures is to provide a mathematical description of the quantum formalism in finite dimension and to introduce the mathematical concepts and tools required for the analysis of such quantum systems and their dynamics. On the one hand, we will study the key aspects of quantum information theory. On the other hand, we will describe certain properties of quantum dynamics that need to be taken into account in the implementation of quantum algorithms and that will be applied to emblematic systems. The interaction with an external classical electromagnetic field will also be considered both from a theoretical and a numerical point of view.

Visit the full page about this course

Visit the full page about this course

This is the master thesis project.

Visit the full page about this course