40 Postdoctoral Fellowship at Inria, France

Non-Linear Equations are ubiquitous in domains such as geometric modelling, scientific computing, data analysis, . . . . Finding their solutions is an instance of non-linear optimisation problems. A classical approach to tackle these non-linear problems is to use local methods, such as Newton method or gradient descent methods. This family of methods is widely used in practice, but it only provides local extrema with no guaranty for a global solution. Moreover, it is very sensitive to the choice of the initial point.

The CLAIRE project focuses on the identification of molecular mechanisms associated with drug-tolerance. Despite the encouraging outcomes observed in cancer patients treated with targeted therapies, tumor responses are only partial and are followed by acquired resistance. It is thought that resistance emerges from drug-tolerant cells that survive treatment through an adaptive response, that does not involve genetic resistance mutations. These drug-tolerant cell populations are considered the main drivers of the partial tumor response observed in patients. In this context, oncogene-driven lung cancer is a paradigm system of the success and limitations of targeted therapy, and thus makes for a relevant system to identify molecular regulatory mechanisms of adaptive survival to therapy.

Candidates for postdoctoral positions are recruited after completing their Ph.D. or after a first postdoctoral period: for the candidates who obtained their Ph.D. in the Northern hemisphere, the date of the Ph.D. defence shall be later than 1 September 2021; in the Southern hemisphere, later than 1 April 2021.

Candidates for postdoctoral positions are recruited after the end of their PhD or after a first post-doctoral period: for the candidates who obtained their PhD in the Northern hemisphere, the date of the defense shall be later than 1 September 2021; in the Southern hemisphere, later than 1 April 2021.

Extreme value theory is a branch of probability and statistics dealing specifically with the behaviour of a probability distribution in its tails. More precisely, it focuses on the asymptotic behaviour of the largest, or lowest, observations in a collection of random observations from the same distribution. In extreme value statistics, the main problems are typically the estimation of the extreme value index and extreme quantiles associated to a random variable of interest Y with unknown underlying distribution. The extreme value index drives the behaviour of the distribution of Y in its right tail. We refer to [1,2] for a general account on extreme value statistics.

Surface electromyography (EMG) is the non-invasive recording of electrical activity produced by skeletal muscle contractions. Surface electroencephalography (EEG) is the recording of electrical activity produced by the activity of neural assemblies in the brain. In both modalities, there is a strong interest in recovering the underlying physiological sources, either the motor unit spike trains or the activity of cortical neural assemblies, that explain the recordings. The EEG community has long recognized the importance of forward modelling. As a consequence, the majority of approaches to estimate cortical sources from EEG measurements rely on knowledge of the volume conductor, whose geometry is provided by magnetic resonance imaging (MRI). However, an important limitation of these approaches is our imperfect knowledge of the tissue properties, in particular their conductivities, used to construct the model. In contrast to EEG, it is possible to implant electrodes inside the volume conductor during EMG acquisitions. This procedure provides us with a unique opportunity to estimate tissue conductivities in vivo and minimally invasively.

Nowadays, there are many scientific fields where the need for computing power and data processing capacity goes beyond what current machines can provide. In radio astronomy, for example, the international SKA project aims to create the largest telescope in the world in order to observe a part of the Universe. A very large volume of data is generated at the telescope level and then transits to geo-distributed data centers to be pre-processed (filtering, reduction) in real time at a rate of 10TB/s. The output data is then sent to a supercomputer to be saved and fed into numerical simulations. At this stage, the computing power and storage resources required are such that machines capable of reaching the exascale become necessary. To date, only a few supercomputers such as Frontier at Oak Ridge National Laboratory (USA) have this capability, but in the coming months, new systems will be deployed. However, the efficient use of these systems raises new challenges, especially regarding data management.

The SISTM (Statistics in Systems and Translational Medicine) team directed by Pr. Rodolphe Thiébaut is involved in PREVAC and PREVAC–UP and had been also involved in many international consortium for EBOLA, ie. EBOVAC1 and EBOVAC3 and coordinates EBOVAC2. In this context one of its main goal is to conduct the analysis and the modelling of the immune response to the Ebola vaccine strategy, using data including immunological substudies recording many biomarkers (cell phenotype, functionality, gene expression, antibody titers…).

Enabling the Internet accessibility in mobility, in particular on board of public land and air transport (trains, buses, airliners), as well as the access to secure communication servers (for combat aircraft or military vehicles, etc.)requires the development of agile, ultra-flat and low-cost Ka-band antennas. Nonetheless, meeting the targeted performance while ensuring a low-profile and cost-effective antenna is a major challenge. To address it, the antenna architecture proposed in this project consists of a planar radiating panel associated with two compact deflectors, whose rotation ensures the beam-steering capacity. Moreover, the compactness and the moderate cost of the deflectors can be obtained thanks to the subwavelength structuration of a dielectric matter by usingadditive manufacturing. Indeed, the sub-wavelength patterning technique has recently shown the possibility to realize antenna components much thinner than their homogeneous bulk counterparts, with equivalent or even better radiofrequency performance.

When a person has their hands busy performing a task like driving a car or piloting an airplane, voice is a fast and efficient way to achieve interaction. In aeronautical communications, the English language is most often compulsory. Unfortunately, a large part of the pilots are not native English and speak with an accent dependent on their native language and are therefore influenced by the pronunciation mechanisms of this language. Inside an aircraft cockpit, non-native voice of the pilots and the surrounding noises are the most difficult challenges to overcome in order to have efficient automatic speech recognition (ASR). The problems of non-native speech are numerous: incorrect or approximate pronunciations, errors of agreement in gender and number, use of non-existent words, missing articles, grammatically incorrect sentences, etc. The acoustic environment adds a disturbing component to the speech signal. Much of the success of speech recognition relies on the ability to take into account different accents and ambient noises into the models used by ARP.

Cybersecurity is a major concern everywhere with the growth of connected devices that are beyond common computers. To circumvent these problems, decades of research and development have led to build new techniques and tools to fight back against the attacks on the Internet. Nonetheless, the number of attacks and their magnitude still grow. The attack surface continues to increase along with the number of connected devices but also due to the number of applications, services or software that today make the IT ecosystem far from its origin.

The post-doc will be recruited within the Idefix Team at Inria Saclay, France and will work on a joint project between the Idefix Team and University College London, UK and Cardiff University, UK.  The three supervisors are

Candidates for postdoctoral positions are recruited after the end of their Ph.D. or after a first post-doctoral period: for the candidates who obtained their PhD in the Northern hemisphere, the date of the Ph.D. defense shall be later than 1 September 2021; in the Southern hemisphere, later than 1 April 2021.

Candidates for postdoctoral positions are recruited after the end of their Ph.D. or after a first post-doctoral period: for the candidates who obtained their PhD in the Northern hemisphere, the date of the Ph.D. defense shall be later than 1 September 2021; in the Southern hemisphere, later than 1 April 2021.

The postdoctoral fellow will be recruited at the Inria center of the University of Bordeaux (AUCTUS team), where he/she will start the research project. Two six-month exchange periods will be planned at KAIST (IRiS lab) to share the works and progresses.

 

(18) Postdoctoral Fellowship Position

Postdoc Fellowship Position summary/title: Post-Doctoral Research Visit F/M Reproducible Science and Software Variability

Deadline :2023-06-18

View details & Apply

 

Candidates for postdoctoral positions are recruited after the end of their Ph.D. or after a first post-doctoral period: for the candidates who obtained their PhD in the Northern hemisphere, the date of the Ph.D. defense shall be later than 1 September 2021; in the Southern hemisphere, later than 1 April 2021.

Candidates for postdoctoral positions are recruited after the end of their  Ph.D. or after a first post-doctoral period: for the candidates who obtained their PhD in the Northern hemisphere, the date of the Ph.D. defense shall be later than 1 September 2021; in the Southern hemisphere, later than 1 April 2021.

Multivariate time series refers to a type of data that consists of multiple time series variables observed simultaneously over time. In other words, it consists of a set of observations over time such that each observation has multiple variables measured at the same time. This type of time series can arise in several application domains, such as medical monitoring, environmental sciences such as climate modeling, traffic flow prediction, finance, etc. One of the main challenges in analyzing multivariate time series data is the complexity of the data structure, as there can be many variables that interact with each other in complicated ways over time. This can make it difficult to identify patterns and relationships within the data, as well as to develop accurate prediction models. Anomalies in time series data are defined as patterns that do not conform to the expected behavior of the underlying application. The objective of this research direction is to find unusual events or behaviors in multivariate time series data. It can be useful, for example, to identify abnormal biodiversity phenomena, unusual changes in vegetation or animal populations, or anomalies in climate data. There are several existing methods for multivariate time series anomaly detection, including statistical methods like PCA and ARIMA, e.g., the well-known Matrix Profile problem. However, current MP tools cannot scale to very large time series databases.

Candidates for postdoctoral positions are recruited after the end of their Ph.D. or after a first post-doctoral period: for the candidates who obtained their PhD in the Northern hemisphere, the date of the Ph.D. defense shall be later than 1 September 2021; in the Southern hemisphere, later than 1 April 2021.

As a member of the ANR project, the recruited Postdoctoral researcher will focus more specifically on the computational efficiency of the FEM simulation. Indeed, that kind of simulation involving several organs in contact is typically expensive for several reasons:

In this Postdoc, we want to study active strategies of data acquisition involving a master robot (the ground vehicle) and two slave robots (aerial robots) in order to optimize in terms of accuracy and stability the mapping of the environment. Each robot will be equipped with its own large field vision with a time stamping. We assume also that synchronized velocity control inputs can be sent to the ground and aerial robots. The master robot centralizes the information and manages the multi-robots formation. One robot of the formation, so-called the leader (not necessary the master) is manually driven by a human operator who ensures the completness of the exploration of the scene.

Le premier problème à étudier est celui de la stratégie à adopter pendant le jeu. La stratégie consiste à définir “l’ensemble d’actions coordonnées, d’opérations habiles, de manœuvres en vue d’atteindre un objectif précis”. Dans ce cadre, les principales problématiques scientifiques sont (i) comment formaliser le problème (souvent comme optimisation d’une fonction de coût) et (ii) comment pouvoir définir plusieurs stratégies possibles tout en gardant les mêmes outils pour la mise en œuvre (tactique).

Within the Ofast3D exploratory project, we aim to increase the production capacity of players using 3D printing by fused filament deposition without modifying their infrastructures. This project is a collaboration between the PACAP team for the compilation and optimization aspects for performance, MFX for its experience in 3D printing and MimeTIC for the planning and scheduling aspects.

This postdoc project aims at developing a statistical learning method for the detection and classification of circulating tumor cells in the bloodstream based on their biophysical characteristics. This proposal is part of a innovative approach for the diagnosis and evaluation of cancer metastasis from blood samples.

The Chemical Master Equation (CME) is a linear differential system which describes the progres of a set of chemical reactions. When only a low number of molecules is involved, the occurence of chemical reactions is inherently stochastic and can be modeled with a continuous-time Markov chain (CTMC) [1].

L’ objectif du postdoc est d’obtenir des résultats de contrôle sur un système d’interaction fluide-structure dans le cas où la structure correspond à une paroi élastique localisée sur le bord du domaine du fluide et où le fluide est visqueux et incompressible. Ce type de problème d’interaction entre une structure déformable et un fluide correspond à un système d’équations aux dérivées partielles non linéaires avec une frontière libre. Il existe très peu de résultats de contrôle pour ce système et ils ont été obtenus dans le cas d’une géométrie particulière pour lesquelles il est possible de découpler les variables spatiales. Cela permet notamment d’utiliser une condition au bord plus simple à traiter et de pouvoir séparer simplement les hautes fréquences et les basses fréquences. Dans le cas général, il faudra essayer de retrouver cette propriété des variables spatiales localement près du bord et utiliser des outils d’analyse microlocale pour pouvoir séparer haute et basse fréquences. 

Within the framework of a partnership with team INSERM 1114 in Strasbourg, a software package will be developed which permits to search for an optimal set of experimental parameters in psycho-physical experiments. These parameters are optimal in the sense that they permit optimal achievements, such as fastest reaction times or lowest error rates.

Récemment, il a été démontré dans un cas simple qu’il existe des phénomènes de cut-off pour ces modèles : la distance à l’équilibre reste grande pendant un temps long puis, soudainement, la distribution du processus se rapporche à très grande vitesse de sa valeur d’équilibre. Les méthodes utilisées pour ces résultats sont spécifiques à un cas relativement simple et en espace d’états finis, la question de sa généralisation reste ouverte.

This Postdoctoral Research Fellowship position is proposed by the Mathrisk INRIA  Research team, https://team.inria.fr/mathrisk/en/, which is common with INRIA, the Ecole Nationale des Ponts et Chaussées and the University Gustave Eiffel. 

The  Mathrisk team addresses broad research topics in the area of  mathematical handling of risk, embracing risk measurement and risk management, modeling and optimization in quantitative finance and other related domains where risk control is paramount. It also develops a numerical platform for quantitative finance(http://www.premia.fr), supported by  a consortium of financial institutions. Mathematical expertise of the team include stochastic modeling, stochastic analysis, in particular stochastic (partial) differential equations and various aspects of stochastic control and optimal stopping of these equations, stochastic optimization, dynamic game theory, random graphs, martingale optimal transport and numerical probability.

The Cambium team at Inria Paris conducts research on programming languages, including programming language design and implementation; semantics and type systems; and program verification. The team is known as the home of the OCaml programming language, of the CompCert verified compiler, and of the diy tool suite. The team comprises 6 permanent researchers and (as of spring 2023) 5 PhD students.

In the area of program verification, we are expert users of the Coq proof assistant and of the Iris program verification framework. Over the years, we have proposed several extensions of Iris, including time credits (to reason about amortized asymptotic time complexity), space credits (to reason about space complexity in the presence of garbage collection), and views and subjective assertions (to reason about OCaml 5’s weak memory model). Two of our ongoing projects in this area are Gospel, which aims to develop a user-friendly specification language for OCaml, and Osiris, which aims to develop a ready-to-use instance of Iris with support for a large subset of OCaml.

The MiMove team at Inria Paris (https://mimove.inria.fr) undertakes research enabling next-generation mobile distributed systems, from their conception and design to their runtime support, with focus on both middleware and data.

MiMove’s research, in particular, accounts for the presence of large and volatile populations of mobile and resource-constrained devices, such as user handhelds and wireless sensors, as well as of fast evolving, global networking and computing infrastructures. We study emergent systems that are dynamically composed from heterogeneous networked resources in the near (IoT, edge) and remote (fog, cloud) environment, and that furthermore adapt to their highly changing execution context, whether technical or physical. 

Synchronization primitives such as barriers and locks are the building blocks of multithreaded applications on multicore architectures: improving their performance has an impact on the performance and scalability of the majority of today’s applications [1]. Meanwhile, the Linux scheduler must solve a complex problem that has both a temporal and a spatial dimension: it has to make very fast decisions at a high frequency in order to decide when and where to place each thread—and has been shown to perform poorly at this task [2]. On modern architectures with a large number of cores, the overhead of synchronization is often a significant bottleneck. While previous research heavily focused on improving the performance of synchronization primitives in multithreaded applications, it was conducted in isolation from research on scheduling algorithms, as the scheduler traditionally considers applications as black boxes, including the synchronization code they use.