Search

We wish to define the limits of a classical classification model based on deep learning when applied to abstract images, which do not represent visually identifiable objects.QR codes (Quick Response codes) fall into this category of abstract images: one bit corresponding to one encoded character, QR codes were not designed to be decoded manually. To understand the limitations of a deep learning-based model for abstract image classification, we train an image classification model on QR codes generated from information obtained when reading a health pass. We compare a classification model with a classical (deterministic) decoding method in the presence of noise. This study allows us to conclude that a model based on deep learning can be relevant for the understanding of abstract images., in French language. RFIAP 2022 - Reconnaissance des Formes, Image, Apprentissage et Perception, Jul 2022, Vannes (Bretagne), France

This article presents the methods used to parallelize a new computer vision application. The system is able to automatically detect meteor from non-stabilized cameras and noisy video sequences. The application is designed to be embedded in weather balloons or for airborne observation campaigns. Thus, the final target is a low power system-on-chip (< 10 Watts) while the software needs to compute a stream of frames in real-time (> 25 frames per second). For this, first the application is split in a tasks graph, then different parallelization techniques are applied. Experiment results demonstrate the efficiency of the parallelization methods. For instance, on the Raspberry Pi 4 and on a HD video sequence, the processing chain reaches 42 frames per second while it only consumes 6 Watts., in French language, COMPAS 2023 - Conf{\'e}rence francophone d'informatique en Parall{\'e}lisme, Architecture et Syst{\`e}me, Jul 2023, Annecy (France), France

"Data is the new oil", in short, data would be the essential source of the ongoing fourth industrial revolution, which has led some commentators to assimilate too quickly the quantity of data to a source of wealth in itself, and consider the development of big data as an quasi direct cause of profit. Human resources management is not escaping this trend, and the accumulation of large amounts of data on employees is perceived by some entrepreneurs as a necessary and sufficient condition for the construction of predictive models of complex work behaviors such as absenteeism or job performance. In fact, the analogy is somewhat misleading: unlike oil, there are no major issues here concerning the production of data (whose flows are generated continuously and at low cost by various information systems), but rather their ''refining'', i.e. the operations necessary to transform this data into a useful product, namely into knowledge. This transformation is where the methodological challenges of data valuation lie, both for practitioners and for academic researchers. Considerations on the methods applicable to take advantage of the possibilities offered by these massive data are relatively recent, and often highlight the disruptive aspect of the current ''data deluge'' to point out that this evolution would be the source of a revival of empiricism in a ''fourth paradigm'' based on the intensive and ''agnostic'' exploitation of massive amounts of data in order to bring out new knowledge, following a purely inductive logic. Although we do not adopt this speculative point of view, it is clear that data-driven approaches are scarce in quantitative HRM studies. However, there are well-established methods, particularly in the field of data mining, which are based on inductive approaches. This area of quantitative analysis with an inductive aim is still relatively unexplored in HRM ( apart from typological analyses). The objective of this paper is first to give an overview of data driven methods that can be used for HRM research, before proposing an empirical illustration which consists in an exploratory research combining a latent profile analysis and an exploration by Gaussian graphical models., in French Language. 33{\`e}me congr{\`e}s de l'AGRH (association francophone de gestion des resources humaines), Unversit{\'e} de Bretagne Occidentale (UBO), Oct 2022, Brest, France

Logs are a fundamental component of modern computer systems. They enable the analysis and monitoring teams to understand any abnormal or malicious behavior that may have occurred. The continuous increase in the volume of logs generated by these systems made it unsuitable for manual inspection and represents a real challenge with regard to process automation. In order to process these data, several log-structuring solutions have been developed. In this article, we analyze the capabilities of two solutions in order to meet the challenges of Cloud Computing in terms of efficiency and effectiveness. Our work focuses on the impact of parameterization and preprocessing on the performance of these methods -- two important steps as they require human intervention, which is incompatible with with the automation of the log-structuring process., in French language, Extraction et Gestion des Connaissances (EGC), 2021

Logs record valuable system information at runtime. They are widely used by data-driven approaches for development and monitoring purposes. Parsing log messages to structure their format is a classic preliminary step for log-mining tasks. As they appear upstream, parsing operations can become a processing time bottleneck for downstream applications. The quality of parsing also has a direct influence on their efficiency. Here, we propose USTEP, an online log parsing method based on an evolving tree structure. Evaluation results on a wide panel of datasets coming from different real-world systems demonstrate USTEP superiority in terms of both effectiveness and robustness when compared to other online methods., in French language, Extraction et Gestion des Connaissances: EGC'2022, 2022

Chjara, breeder in Carg{\`e}se, has n wild pigs. She would like to sort her herd by weight to better meet the demands of her buyers. Each beast has a distinct weight, alas unknown to Chjara. All she has at her disposal is a Roberval scale, which allows her to compare two pigs only at the cost of an acrobatic manoeuvre. The balance, quite old, can break at any time. Chjara therefore wants to sort his herd in a minimum of weighings, but also to have a good estimate of the result after each weighing.To help Chjara, we pose the problem of finding a good anytime sorting algorithm, in the sense of Kendall's tau distance between provisional result and perfectly sorted list, and we bring the following contributions:- We introduce Corsort, a family of anytime sorting algorithms based on estimators.- By simulation, we show that a well-configured Corsort has a near-optimal termination time, and provides better intermediate estimates than the best sorting algorithms we are aware of., Comment: in French language, AlgoTel 2023 - 25{\`e}mes Rencontres Francophones sur les Aspects Algorithmiques des T{\'e}l{\'e}communications, May 2023, Cargese, France

Artificial Intelligence is more and more pervasive in our lives. Many important decisions are delegated to AI algorithms: accessing higher education, determining prison sentences, autonomously driving vehicles... Engineers and researchers are educated to this field, while the general population has very little knowledge about AI. As a result, they are very sensitive to the (more or less accurate) ideas disseminated by the media: an AI that is unbiased, infallible, and will either save the world or lead to its demise. We therefore believe, as highlighted by UNESCO, that it is essential to provide the population with a general understanding of AI algorithms, so that they can choose wisely whether to use them (or not). To this end, we propose a serious game in the form of a civic debate aiming at selecting an AI solution to control a pandemic. This game is targeted at high school students, it was first experimented during a science fair, and is now available freely., in French language

Research activities are generally observed and evaluated through the prism of their production and financial elements or team composition. In addition to standardized management indicators and bibliometrics, the French National Research Institute for Sustainable Development (IRD) has been building new indicators for the last ten years, based on the annual regulatory declarations of the Institute's researchers. Different quality management tools allow the evolution of the different interfaces. This source of data, more ''open'' and more ''useful'' through its integration into the Institute's information system, is adapted to the needs of the multi-year management of research at the IRD. The aim is twofold: (1) to make progress in the evaluation of research and in the mastery of information by all actors, (2) to enlighten as many actors as possible via more efficient digital circuits and tools. The purpose of this article is to explain how the IRD is changing the entire production chain and the indicators of researchers' activities to better map scientific activities., in French language

My research-creation process coincides with the encounter with the ''digital paradigm'' and the attempt to incorporate it into the foundation of my scenic writing. I propose in this paper to give an account from a director point of view of how I realized my shows between 2000 and 2007 and which researches influenced the process. I will formulate some remarks on the fragilities that can arise in a ''technological laboratory of staging''., in French language

Low information transfer rate is a major bottleneck for brain-computer interfaces based on non-invasive electroencephalography (EEG) for clinical applications. This led to the development of more robust and accurate classifiers. In this study, we investigate the performance of quantum-enhanced support vector classifier (QSVC). Training (predicting) balanced accuracy of QSVC was 83.17 (50.25) %. This result shows that the classifier was able to learn from EEG data, but that more research is required to obtain higher predicting accuracy. This could be achieved by a better configuration of the classifier, such as increasing the number of shots., in French language

The IRMA project aims to design innovative methodologies for research in the field of historical and archaeological heritage based on a combination of medical imaging technologies and interactive 3D restitution modalities (virtual reality, augmented reality, haptics, additive manufacturing). These tools are based on recent research results from a collaboration between IRISA, Inrap and the company Image ET and are intended for cultural heritage professionals such as museums, curators, restorers and archaeologists., in French language

Since its conception, the computer has found applications to accompany human creativity. Today, the debate about computers and creativity involves several challenges, such as understanding human creativity, modeling the creative process, and programming the computer to exhibit behavior that appears to be creative to some extent. In this paper, we are interested in how the computer can be used as a tool to promote creativity in a musical composition. We automatically extracted the brightness curve from a silent movie and then used it to compose a piece of music to accompany the movie. We extracted several parameters from the brightness curve, and applied compositional rules from these parameters to write the instrumental music for the film. The final composition has a synchronicity and aesthetic fit with the film that are surprising. This compositional process also allowed for a degree of aesthetic freedom that would otherwise have been impossible., in French language

We consider the problem of phase retrieval that consists in recovering an n-dimensional real vector from the magnitude of its m-linear measurements. This paper presents a new approach allowing to lift the classical global Lipschitz continuity requirement on the gradient of the non-convex objective to minimize. We propose a mirror descent algorithm based on a wisely chosen Bregman divergence. We show that when the number of measurements m is large enough, the mirror descent algorithm, carefully initialized, converges linearly with a dimension-independent convergence rate. Consequently, the original signal can be reconstructed exactly up to a global sign change. We state our results for two types of measurements: iid standard Gaussian and those obtained by Coded Diffraction Patterns (CDP) for Randomized Fourier Transform.; Dans ce travail, nous considérons le problème de reconstruction de phase, qui consiste à recouvrer un signal de dimension n à partir des modules de ses m mesures linéaires. Nous présentons une nouvelle approche permettant de se passer de la Lipschitz continuité globale du gradient de l'objective non-convexe à minimiser. Pour ce faire, nous introduisons un algorithme de descente miroir non-euclidienne via la divergence de Bregman d'une entropie bien choisie. Nous montrons que lorsqu'on le nombre de mesures m est suffisamment grand, alors avec grande probabilité, l'algorithme de descente miroir, bien initialisé, converge linéairement avec un taux indépendant de la dimension n, permettant ainsi de recouvrer le signal original à un changement de signe près. Nous énonçons nos résultats pour deux types de mesures: Gaussiennes iid et celles obtenues par des masques codés pour des transformées de Fourier aléatoires.

LibStaffer is a staff scheduling tool for libraries provided by Springshare. In this article, we present the analysis that we implemented to determine the potential of LibStaffer in order to simplify staff scheduling in library branches with various configurations, as well as in transversal units with a focus on public services. As a starting point, we focused on the questions available in Philippe Lenepveu and Marc Maisonneuve's work on scheduling tools for libraries. We then enriched the initial list with questions that emerged among several colleagues. After a two-month LibStaffer trial period, we were able to answer those questions and were convinced to take out a subscription to the tool. -- LibStaffer est un outil de gestion de planning de service pour les biblioth\`eques propos\'e par Springshare. Dans cet article, nous exposons l'analyse que nous avons mise en {\oe}uvre pour d\'eterminer le potentiel de LibStaffer en mati\`ere de simplification de la gestion de planning d'accueil, sur des implantations de configurations diverses, ainsi que d'activit\'es transversales de service au public. Nous nous sommes fond\'es sur les questions \'etablies par Philippe Lenepveu et Marc Maisonneuve, dans leur ouvrage consacr\'e aux logiciels de gestion de planning pour les biblioth\`eques, et les avons enrichies par les interrogations de plusieurs coll\`egues. B\'en\'eficier d'un test de LibStaffer, de pr\`es de deux mois, nous a permis de r\'epondre \`a ces questions et nous a convaincus de prendre une souscription \`a l'outil., Comment: 17 pages, in French

In this article, we develop a deterministic compartmental model including the spatio-temporal aspect and the wearing of masks by the general public for the spread of the Covid-19 epidemic. This model is based on the SIR model of Kermack and McKendrick and includes mobility terms that model the movement of individuals wearing masks between Guadeloupe andMartinique. The objective of the model is to analyze the effects of mobility and the wearing of masks by the general public on the spread of the Covid-19 epidemic in these two regions.Numerical simulations of this model show the importance of wearing masks by the general public, even if they are not very inefficient, and indicate that the model is capable of qualitatively simulating the propagation trends of the Covid-19 epidemic. This work is part ofmy doctoral thesis, the subject of which is entitled {\guillemotleft} stochastic modeling of Covid-19 in an island environment taking into account the spatio-temporal aspect {\guillemotright}., Comment: in French language

Spoken medical dialogue systems are increasingly attracting interest to enhance access to healthcare services and improve quality and traceability of patient care. In this paper, we focus on medical drug prescriptions acquired on smartphones through spoken dialogue. Such systems would facilitate the traceability of care and would free clinicians' time. However, there is a lack of speech corpora to develop such systems since most of the related corpora are in text form and in English. To facilitate the research and development of spoken medical dialogue systems, we present, to the best of our knowledge, the first spoken medical drug prescriptions corpus, named PxSLU. It contains 4 hours of transcribed and annotated dialogues of drug prescriptions in French acquired through an experiment with 55 participants experts and non-experts in prescriptions. We also present some experiments that demonstrate the interest of this corpus for the evaluation and development of medical dialogue systems., Ali Can Kocabiyikoglu,Fran\c{c}ois Portet, Prudence Gibert, Herv\'e Blanchon, Jean-Marc Babouchkine, Ga\"etan Gavazzi. A Spoken Drug Prescription Dataset in French for Spoken Language Understanding. LREC2022, Marseille, France, 21-22-23 June 2022

With the popularity of virtual assistants (e.g., Siri, Alexa), the use of speech recognition is now becoming more and more widespread.However, speech signals contain a lot of sensitive information, such as the speaker's identity, which raises privacy concerns.The presented experiments show that the representations extracted by the deep layers of speech recognition networks contain speaker information.This paper aims to produce an anonymous representation while preserving speech recognition performance.To this end, we propose to use vector quantization to constrain the representation space and induce the network to suppress the speaker identity.The choice of the quantization dictionary size allows to configure the trade-off between utility (speech recognition) and privacy (speaker identity concealment)., Journ{\'e}es d'{\'E}tudes sur la Parole - JEP2022, Jun 2022, {\^I}le de Noirmoutier, France

Modern Natural Language Processing (NLP) models based on Transformer structures represent the state of the art in terms of performance on very diverse tasks. However, these models are complex and represent several hundred million parameters for the smallest of them. This may hinder their adoption at the industrial level, making it difficult to scale up to a reasonable infrastructure and/or to comply with societal and environmental responsibilities. To this end, we present in this paper a model that drastically reduces the computational cost of a well-known French model (CamemBERT), while preserving good performance., in French language. CAp (Conf{\'e}rence sur l'Apprentissage automatique), Jul 2022, Vannes, France

The paper offers a contribution to the interdisciplinary constructs of analyzing fairness issues in automatic algorithmic decisions. Section 1 shows that technical choices in supervised learning have social implications that need to be considered. Section 2 proposes a contextual approach to the issue of unintended group discrimination, i.e. decision rules that are facially neutral but generate disproportionate impacts across social groups (e.g., gender, race or ethnicity). The contextualization will focus on the legal systems of the United States on the one hand and Europe on the other. In particular, legislation and case law tend to promote different standards of fairness on both sides of the Atlantic. Section 3 is devoted to the explainability of algorithmic decisions; it will confront and attempt to cross-reference legal concepts (in European and French law) with technical concepts and will highlight the plurality, even polysemy, of European and French legal texts relating to the explicability of algorithmic decisions. The conclusion proposes directions for further research, Version 1.0-6 mai 2022 RESUME L'article propose une contribution aux constructions interdisciplinaires de l'analyse des enjeux d'équité dans les décisions algorithmiques automatiques. La section 1 montre que les choix techniques en apprentissage supervisé ont des implications sociales dont il faut prendre la mesure. La section 2 propose une approche contextuelle de la question de la discrimination de groupe non intentionnelle, c'est-à-dire de règles de décision facialement neutres mais qui génèrent des impacts disproportionnés selon les groupes sociaux (selon les cas : genrés, raciaux ou ethniques). La contextualisation portera sur les systèmes juridiques des États-Unis d'un côté, de l'Europe d'un autre côté. En particulier, la législation et la jurisprudence tendent à promouvoir des critères d'équité différents de part et d'autre de l'Atlantique. La section 3 est consacrée à l'explicabilité des décisions algorithmiques ; elle confrontera et tentera de croiser les concepts juridiques (en droit européen et en droit français) avec les concepts techniques et mettra en exergue la pluralité, voire la polysémie, des textes juridiques européens et français relatifs à l'explicabilité des décisions algorithmiques. La conclusion propose des orientations pour la recherche.

C'est une belle histoire !, This is lmy paper !

The textual content of a document and its publication date are intertwined. For example, the publication of a news article on a topic is influenced by previous publications on similar issues, according to underlying temporal dynamics. However, it can be challenging to retrieve meaningful information when textual information conveys little. Furthermore, the textual content of a document is not always correlated to its temporal dynamics. We develop a method to create clusters of textual documents according to both their content and publication time, the Powered Dirichlet-Hawkes process (PDHP). PDHP yields significantly better results than state-of-the-art models when temporal information or textual content is weakly informative. PDHP also alleviates the hypothesis that textual content and temporal dynamics are perfectly correlated. We demonstrate that PDHP generalizes previous work --such as DHP and UP. Finally, we illustrate a possible application using a real-world dataset from Reddit., Comment: in French

In the PRIM project, we aim at giving people the power to create scenagrams (interaction scenarios between a human and digital devices) without the need to learn programming or to ask for computer scientists. In this project, software design follows an unconventional approach, far from classical codes, to embody human thinking (based on interactions) instead of computer logic (based on algorithms). The main idea rests on a new time representation using a PRIM-specific timeline instead of a standardized timeline. We evaluated acceptability and cognitive compatibility of this new timeline with 50 participants. Results are very promising. In this paper, we detail qualitative evaluation results about the interest of such software in the field of disability., Avec le projet PRIM, nous souhaitons permettre à des personnes de créer des scénagrammes (scénarios d'interaction entre un humain et des dispositifs numériques) sans devoir se former à la programmation ou faire appel à des informaticiens. Dans PRIM, la conception du logiciel renverse les codes classiques dans le sens où il représente la pensée de l'humain (fondée sur les interactions) plutôt que la logique de l'ordinateur (fondée sur l'algorithmique) grâce à une représentation renouvelée du temps, abandonnant la timeline normée pour une timeline modifiée, propre à PRIM. Après avoir évalué l'acceptabilité et la compatibilité cognitive de cette timeline par 50 participants et obtenu des résultats encourageants, nous présentons, dans ce papier, les résultats de l'évaluation qualitative de l'intérêt d'un tel outil pour les participants et, plus spécifiquement, dans le domaine du handicap.

Explanation methods have emerged as an important tool to highlight the features responsible for the predictions of neural networks. There is mounting evidence that many explanation methods are rather unreliable and susceptible to malicious manipulations. In this paper, we particularly aim to understand the robustness of explanation methods in the context of text modality. We provide initial insights and results towards devising a successful adversarial attack against text explanations. To our knowledge, this is the first attempt to evaluate the adversarial robustness of an explanation method. Our experiments show the explanation method can be largely disturbed for up to 86% of the tested samples with small changes in the input sentence and its semantics.

In this article, we present the SILKNOW project (Silk heritage in the Knowledge Society: from punched card to Big Data, Deep Learning and visual/tangible simulations) (2018-2021). This project aimed to use Semantic Web technologies to give greater visibility to silk objects produced and consumed in Europe between the 15th and 19th centuries. Silk is a particularly important material in European history, and it has produced some exceptional objects of great historical interest. However, it is a threatened heritage that is little known to the general public. We show the interest of using Semantic Web technologies to give more visibility to such a heritage, by describing the results we have obtained. We present the methodology used to develop a knowledge graph, and in particular the different steps that were necessary to create the underlying data model, based on the CIDOC CRM or CIDOC Conceptual Reference Model. We also propose a CIDOC CRM-compatible extension to express the complex semantics of the creation and production process of ancient silk fabrics., in French

This article shows that the boundaries between the editing and online publishingprofessions are losing their strength. In this context it would only make sense that the wayhypertexts are documented be renewed, especially facing of the Web's evolution. We arethinking in particular of the trickier scholar hypertexts documentation process - specifically inscientific or cultural contexts. The purpose of this article is to demonstrate that, consideringthe numerous branches of the Web, the hypertext enhance of a document of quality can onlybe done through a proper dialogue between authors, editors, and broadcasters. It would satisfythe readership as they could reach the appropriate information. It will also be shown that eachactor in this auctorial-editorial process would be a gainer. Indeed, a qualitative formalizationwork would be coupled with a strong broadcasting scope. Finally, we will point out that thiswork of mediating must be led by an actor of information-communication, to make the textunderstandable to both humans and machines. This meditative act is designated here under theterm of serial documentarisation., in French, H2PTM, Oct 2021, Paris, France

Generating user activity is a key capability for both evaluating security monitoring tools as well as improving the credibility of attacker analysis platforms (e.g., honeynets). In this paper, to generate this activity, we instrument each machine by means of an external agent. This agent combines both deterministic and deep learning based methods to adapt to different environment (e.g., multiple OS, software versions, etc.), while maintaining high performances. We also propose conditional text generation models to facilitate the creation of conversations and documents to accelerate the definition of coherent, system-wide, life scenarios., in French

As software has become an integral part of scientific workflows, reproducible research practices must take it into account. In what way? Archiving source code is a necessary but insufficient condition. The ability to redeploy software environments, which at first sight may be viewed as a technical detail, is in fact a requirement. This article explores tools and methods to achieve this goal., in French

International audience

In this paper, we propose a method for coarse camera pose computation which is robust to viewing conditions and does not require a detailed model of the scene. This method meets the growing need of easy deployment of robotics or augmented reality applications in any environments, especially those for which no accurate 3D model nor huge amount of ground truth data are available. It exploits the ability of deep learning techniques to reliably detect objects regardless of viewing conditions. Previous works have also shown that abstracting the geometry of a scene of objects by an ellipsoid cloud allows to compute the camera pose accurately enough for various application needs. Though promising, these approaches use the ellipses fitted to the detection bounding boxes as an approximation of the imaged objects. In this paper, we go one step further and propose a learning-based method which detects improved elliptic approximations of objects which are coherent with the 3D ellipsoid in terms of perspective projection. Experiments prove that the accuracy of the computed pose significantly increases thanks to our method and is more robust to the variability of the boundaries of the detection boxes. This is achieved with very little effort in terms of training data acquisition -- a few hundred calibrated images of which only three need manual object annotation. Code and models are released at https://github.com/zinsmatt/3D-Aware-Ellipses-for-Visual-Localization., Comment: Presented at 3DV 2020. Code and models released at https://github.com/zinsmatt/3D-Aware-Ellipses-for-Visual-Localization

This study provides a large-scale mapping of the French media space using digital methods to estimate political polarization and to study information circuits. We collect data about the production and circulation of online news stories in France over the course of one year, adopting a multi-layer perspective on the media ecosystem. We source our data from websites, Twitter and Facebook. We also identify a certain number of important structural features. A stochastic block model of the hyperlinks structure shows the systematic rejection of counter-informational press in a separate cluster which hardly receives any attention from the mainstream media. Counter-informational sub-spaces are also peripheral on the consumption side. We measure their respective audiences on Twitter and Facebook and do not observe a large discrepancy between both social networks, with counter-information space, far right and far left media gathering limited audiences. Finally, we also measure the ideological distribution of news stories using Twitter data, which also suggests that the French media landscape is quite balanced. We therefore conclude that the French media ecosystem does not suffer from the same level of polarization as the US media ecosystem. The comparison with the American situation also allows us to consolidate a result from studies on disinformation: the polarization of the journalistic space and the circulation of fake news are phenomena that only become more widespread when dominant and influential actors in the political or journalistic space spread topics and dubious content originally circulating in the fringe of the information space., Comment: IC2S2, Jul 2021, Zurich, Switzerland

Sea fishing is a highly mobile activity, favoured by the vastness of the oceans, the absence of physical boundaries and the abstraction of legislative boundaries. Understanding and anticipating this mobility is a major challenge for fisheries management issues, both at the national and international levels. ''FisherMob'' is a free Gama tool designed to study the effect of economic and biological factors on the dynamics of connected fisheries. It incorporate the most important processes involved in fisheries dynamics: fish abundance variability, price of the fishing effort and ex-vessel fish market price that which depends on the ratio between offer and demand. The tool uses as input a scheme of a coastal area with delimited fishing sites, fish biological parameters and fisheries parameters. It runs with a userfriendly graphic interface and generates output files that can be post-processed easily using graphic and statistical software., Comment: in French

We introduce a new formalisation of languages, called keyboards. We consider a set of elementary operations (writing/erasing a letter, going to the right or to the left,...) and we define a keyboard as a set of finite sequences of such operations, called keys. The corresponding language is the set of words obtained by applying some sequence of those keys. Unlike classical models of computation, every key can be applied anytime. We define various classes of languages based on different sets of elementary operations, and compare their expressive powers. We also compare them to well-known classes of languages (Chomsky hierarchy). We obtain a strict hierarchy of languages, whose expressivity is orthogonal to the one of the aforementionned classical models. -- Nous introduisons une nouvelle repr\'esentation de langages, les claviers. On se munit d'un ensemble d'op\'erations \'el\'ementaires (ajout, effacement d'une lettre, d\'eplacement \`a droite, \`a gauche, ...), et on d\'efinit un clavier comme un ensemble de suites finies d'op\'erations \'el\'ementaires, appel\'ees touches. Son langage sera l'ensemble des mots obtenus en appliquant une suite quelconque de touches. Contrairement \`a des mod\`eles de calcul classiques, toutes les touches peuvent \^etre appliqu\'ees \`a tout moment. En premier lieu nous d\'efinissons diff\'erentes classes de claviers en faisant varier l'ensemble des op\'erations \'el\'ementaires autoris\'ees, et nous comparons l'expressivit\'e des classes de langages obtenues. Nous comparons \'egalement ces classes \`a la hi\'erarchie de Chomsky. Nous obtenons que toutes les classes \'etudi\'ees sont diff\'erentes, et nous caract\'erisons les classes inclues dans les rationnels et les alg\'ebriques. L'expressivit\'e des claviers semble orthogonale \`a celle des mod\`eles \'evoqu\'es pr\'ec\'edemment., Comment: Two versions, in French and in English

Our contribution concerns interactive decision support systems for group decision support. Through this study, we apply to implement a decisional process aiming to represent the multiplicity of actors, their diversity, their behaviors and their interactions. In this context, we contribute to the design and development of a group decision support system. The system is modeled by a multi agents system while exploiting a negotiation protocol based on mediation and concession. This protocol allows decision-makers to express their preferences using multicriteria analysis methods, mainly the method by total aggregation AHP (Hierarchical Process Analysis) and the method by partial aggregation PROMETHEE II ., in French

The Multi-armed bandit offer the advantage to learn and exploit the already learnt knowledge at the same time. This capability allows this approach to be applied in different domains, going from clinical trials where the goal is investigating the effects of different experimental treatments while minimizing patient losses, to adaptive routing where the goal is to minimize the delays in a network. This article provides a review of the recent results on applying bandit to real-life scenario and summarize the state of the art for each of these fields. Different techniques has been proposed to solve this problem setting, like epsilon-greedy, Upper confident bound (UCB) and Thompson Sampling (TS). We are showing here how this algorithms were adapted to solve the different problems of exploration exploitation., in French