Archives

PhD in Computational Social Sciences

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : médialab, Sciences Po
Durée : 3 ans
Contact : pedro.ramaciottimorales@sciencespo.fr
Date limite de publication : 2022-03-12

Contexte :
The médialab at Sciences Po is hiring a doctoral researcher to work on the impact of artificial intelligence on socio-political and media dynamics for project “AI-Political Machines”

Sujet :
This position is part of the project AIPM (AI-Political Machines), funded by the McCourt Institute, devoted to conducting research tackling the challenges of the Internet, Artificial Intelligence, and their impact in society. The goal of AIPM is to improve the understanding of how AI systems perceive large social, political, and informational online systems, and the implications are for algorithmic mediation and the impact on social phenomena. Most AI systems mediating digital ecosystem (generating friend and content recommendations, filtering content, and assisting online searches) are trained using digital traces: networks of friends in social platforms (social graphs), content consumption (clicks, views, shares, retweets, etc.), in addition to text and images, to name a few. The goal of the project is to improve the understanding of what these systems are implicitly inferring about users for producing outputs such as recommendations. Are AI systems capable of inadvertently learning political stances of users when computing recommendations? What are the effects of algorithms mediating digital space in opinions and polarization, or agenda setting dynamics? How to leverage knowledge about machine perception of large social systems in designing better AI systems?

Profil du candidat :
The hired doctoral researcher must be able to conduct data analyses of social and media platform data and conduct experiments involving training and deploying machine learning and deep learning algorithms. In particular, the project involves network modelization, and the use of Graph Neural Network and other Recommender System algorithms and methods. The candidate must also have a real interest in learning social sciences and developing team work.
We encourage students with a natural sciences and technology background (e.g., engineering, computer science and mathematics) interested in becoming proficient in social sciences to apply for the position. Applicants with a background in social sciences (e.g., sociology, political science) and strong computational and modelling skills are also encouraged to apply.

Formation et compétences requises :
Experience with ML in Python and data analysis. Interest in expanding research interest towards the frontier between computer sciences and social sciences.

Adresse d’emploi :
1 place St Thomas d’Aquin, Paris, France

Document attaché : 202202170849_Fiche de poste doctorant McCourt AIPM.pdf

Post-Doc position in the research project Ecology and Dependence ECODEP

Offre en lien avec l’Action/le Réseau : – — –/Doctorants

Laboratoire/Entreprise : Laboratory AGM-UMR 8088, CY Cergy Paris Université
Durée : 12 mois
Contact : doukhan@cyu.fr
Date limite de publication : 2022-05-01

Contexte :
In ecology, it is a matter to understand the dynamics and life history of various species through different environments. Indeed, environmental changes can generate rapid changes in the composition of a given population, its length, its phenotypic character or also its genotype distribution.
Demography is generally concerned with predicting human lifespan as well as the population structure with critical involvement in pension systems and public policy decision making. However, these dynamics raise a number of problems to which historical experience does not provide an answer.

Sujet :
The research proposal is to model population growth and to predict biodiversity using innovative stochastic models with a specific focus on ecological problems. The relevant aspects are related to Taylor’s laws [C1998]. In addition, the difficult problems facing marine ecology, in particular those related to the evolution of the environment and its impact on marine species will be of interest. Finally, the applications will be devoted, among others, to the effects of climate change on coral reefs, ecological abundance modelling and the prediction of marine ecosystems [CFM2006], [R2019], [GH2019]. The biostatistical models are also of interest for the project [DFL2017] and they deserve additional attention.
Some of the stylized facts encountered when working with real-world datasets will be enlightened in the postdoc practice. New modelling frameworks for populations dynamics will incorporate, for instance, covariates and we will investigate their statistical properties [GX2017], [KT2019]; see e.g observation models as in [GK2020]. These problems involve isotonic models, parsimony in the presence of non-linearity and non-stationarity [FLN2018], [D2018]. To conclude, causality [W1954] is of importance also to reduce dimensions as noticed in [Z2018]. Publications appeared in the frame of the project ECODEP, see https://doukhan.u-cergy.fr/publications.html.

Profil du candidat :
This post within the EcoDep project https://doukhan.u-cergy.fr/ecodep.html will contribute from both a theoretical and applicative point of view. This postdoc support will consider various issues in Machine Learning for application to ecology data sets and their interpretation. Beyond the standard ML tools, Oracle inequalities, variable selection, individual sequences or model based predictive methods, the project ecodep want to set a special emphasis on IML interpretable Machine Learning to consider issues of high dimensional time series, eg for abundance data sets or for various models with covariates as meteorological data. Specific skills in time series analysis are welcome. IML methods either analyze model components, model sensitivity, or surrogate models.
The post-doc position will be based in Cergy Pontoise, CY Cergy Paris Université, with collaborations in France and abroad for collaborative work within the Ecodep community. English is necessary: a candidate should thus be able to travel by invitations. Travel to Columbia (NYC), Santiago (Chile), Iena (Germany) and several other locations connected to ECODEP are envisioned (depending on the evolution of the pandemic) for research work on:
1- Extensions and applications of Taylor’s law
2- Modeling of abundance
3- Population dynamics
4- Time series issues: isotonicity, causality, covariates, selection
5- Partly observed processes and applications6- Random fields, space time models and their use
7- Panel data studies
8- Risks and data-based studies
Potential locations will depend on the skills and interests in those initial important questions which are not considered in all the Ecodep labs, but CYU will be a fixed point in this position.

Formation et compétences requises :
We are looking for a statistician wishing to be involved in issues and population dynamics This is why we report below several questions of importance in the project. The modelling of population dynamics is of paramount importance in many areas of application.
Qualification PhD in Mathematical Statistics

Adresse d’emploi :
Cergy Pontoise, CY Cergy Paris Université

Document attaché : 202202161046_PageWeb.pdf

Adaptation d’algorithmes de recherche de Process Mining aux besoins d’une startup

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : UTT – LIST3N
Durée : 6 mois
Contact : frederic.bertrand@utt.fr
Date limite de publication : 2022-03-15

Contexte :
Le laboratoire LIST3N (Informatique et Société Numérique) développe des approches efficaces (concepts, modèles, méthodes et outils) pour traiter l’ensemble de la chaîne de traitement des données, des capteurs aux usages, en passant par l’analyse et l’optimisation des données.

Sujet :
Le projet, spécialisé dans le domaine du Process Mining, comprend Frédéric Bertrand, Myriam Maumy, Yoann Valero et Benoit Vuillemin, experts du domaine, et est en collaboration avec la startup Your Data Consulting.

Dans le cadre d’un projet sur le domaine du Process Mining1 en collaboration avec la startup Your Data Consulting, proposant l’outil LiveJourney2, un stagiaire pourrait apporter sa contribution en faisant le lien entre les demandes de l’entreprise et les propositions des travaux de recherche académiques, faites aux travers d’un post doctorat (Benoit Vuillemin) et d’un doctorat (Yoann Valero).
Les travaux attendus incluent, entre autres :
– Étude, amélioration et optimisation des algorithmes de recherche. Cela comprend entre autres, des algorithmes de recherche de règles de prédiction3 et de Deep Learning4. Pour cela, vous serez sous la supervision des concepteurs de ces algorithmes.
– Réunions fréquentes avec les cadres de la startup pour non seulement définir leurs besoins, mais aussi identifier et communiquer ce qui est possible.
– Adaptation et optimisation des algorithmes de recherche aux besoins de l’entreprise.

1 Wil Van Der Aalst, « Process mining », Communications of the ACM, août 2012, https://dl.acm.org/doi/10.1145/2240236.2240257.
2 « Livejourney – Logiciel de Process Mining », s. d., https://www.livejourney.com/fr/.
3 Philippe Fournier-Viger et al., « Mining Partially-Ordered Sequential Rules Common to Multiple Sequences », IEEE Transactions on Knowledge and Data Engineering 27, no 8 (1 août 2015): 2203‑16, https://doi.org/10.1109/TKDE.2015.2405509; Benoit Vuillemin et al., « TSRuleGrowth: Mining Partially-Ordered Prediction Rules From a Time Series of Discrete Elements, Application to a Context of Ambient Intelligence », in Advanced Data Mining and Applications, vol. 11888, Lecture Notes in Computer Science (Cham: Springer International Publishing, 2019), 119‑34, https://doi.org/10.1007/978-3-030-35231-8_9.
4 Leila Arras et al., « Explaining and Interpreting LSTMs », in Explainable AI: Interpreting, Explaining and Visualizing Deep Learning, éd. par Wojciech Samek et al., vol. 11700, Lecture Notes in Computer Science (Cham: Springer International Publishing, 2019), 211‑38, https://doi.org/10.1007/978-3-030-28954-6_11; Antonia Creswell et al., « Generative Adversarial Networks: An Overview », IEEE Signal Processing Magazine 35, no 1 (janvier 2018): 53‑65, https://doi.org/10.1109/MSP.2017.2765202.

Profil du candidat :
Nous avons besoin d’un profil comprenant plusieurs qualités majeures :
– Expérience dans le code, notamment en Python, et ayant envie d’expérimenter de nouveaux langages, tels que Julia,
– Capacité d’identifier des objectifs de haut niveau provenant d’une entreprise et de les matérialiser à l’aide des algorithmes de recherche fournis,
– Force de proposition et de créativité, pour la startup comme pour les chercheurs.

Formation et compétences requises :
BAC +4/+5
Informatique

Adresse d’emploi :
12, rue Marie Curie
10000 Troyes

Document attaché : 202202151246_original.pdf

IOT-ML : Secure Machine Learning on IOT Traces for Daily Activity Discovery

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : Equipe PETRUS INRIA / UVSQ
Durée : 6 mois
Contact : luc.bouganim@inria.fr
Date limite de publication : 2022-06-30

Contexte :
The PETRUS team (Inria/UVSQ), in association with the Hippocad company and the Yvelines Departmental Council, is currently deploying secure home boxes for 10,000 patients in the Yvelines region. These boxes, based on the team’s research results (DBMS embedded in secure hardware), include a personal medical-social database to improve care coordination for dependent people at home. Medical and social workers interact with these secure boxes via a smartphone application. Our objective is to enhance these boxes with the ability to communicate with IOT sensors measuring e.g., luminosity, movement, temperature, to improve patient monitoring. The raw data from the sensors are analyzed by Machine Learning (ML) techniques to identify the patient activities and thus, detect the evolution of patients towards risk situations like depression or illness. Because of their precision, these raw data are however very intrusive. The originality of our approach is to allow a local processing of these data in each box which includes hardware security elements, in order to externalize only the relevant information: alerts, aggregated values on patient dashboards.

Sujet :
ML algorithms build a model based on a training dataset in order to make predictions, in our case, to discover the activity of an individual based on her IOT traces. Beside the classical issues of data representations (from IOT traces to a dataset that can feed an ML algorithm), our approach faces two challenges:
First, we have no possibility to obtain a training dataset for each targeted home-box user. Indeed, we cannot ask elderly people to label their activities during some weeks in order to build the corresponding training datasets: It would be too complex, costly and error prone without a personal assistant. We can however use existing datasets labelled for daily activity discovery (e.g., [1]) and use semi-supervised ML approaches [3] to dynamically adapt the produced model to the targeted home-box user. Indeed semi-supervised approaches use un-labelled data to refine an existing model obtained on labelled data. Other strategies could be defined based on a minimal feedback from the user or on some questionnaires describing the typical activities of the user.
Second, the ML algorithms must be computed inside the home-box, and more precisely in the secure part of the home-box which is composed by a microcontroller with limited RAM resource and a trusted platform module (TPM). Thus the algorithms must be efficient despite limited RAM resources. This may imply to define specific data structures adapted to this environment.

Profil du candidat :
The applicant could be willing to do a Master2 internship or a part-time trainee (Master2 level), or having completed a Master2 and willing to do a PhD

Formation et compétences requises :
• ML algorithm knowledge
• Python (knowledge in C or Rust will be appreciated)

Adresse d’emploi :
UVSQ – Versailles – 45 avenue des états unis.

Document attaché : 202202141426_Master-internship-2022-IOT-ML.pdf

Fouille de modèle pour explorer les avenirs plausibles de la zone des Niayes au Sénégal

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : UMR TETIS
Durée : 6 mois
Contact : camille.jahel@cirad.fr
Date limite de publication : 2022-03-31

Contexte :
La zone des Niayes fournit 70% des produits horticoles à Dakar, profitant d’une nappe phréatique peu profonde, d’un climat favorable et de sols fertiles. Mais ces dernières années ont été marquées par une baisse importante du niveau de la nappe et une salinisation progressive des terres par invasion marine, du fait d’une diminution de la pluviométrie. A cela s’ajoute des problématiques de surexploitation des ressources hydriques par les exploitations agricoles, les agro-industries et les exploitations minières qui ne cessent de s’agrandir. Les prévisions climatiques pour les prochaines années, particulièrement alarmantes pour le Sénégal, tendent à montrer que ces tendances risquent de s’amplifier, menaçant directement toutes les exploitations agricoles de la zone.
Il est donc urgent de prendre la mesure de ces changements pour tenter de les atténuer. Dans ce contexte, une série d’ateliers de prospective ont été menés en 2018, qui ont permis de dessiner les contours de scénarios d’évolution des Niayes (Camara et al., 2020 ). Mais ces scénarios sont dans un registre narratif et qualitatif et doivent maintenant être illustrés d’indicateurs quantitatifs.
Pour cela, une équipe multidisciplinaire de modélisateurs et thématiciens ont écrit un modèle des dynamiques de la zone des Niayes, à l’aide de la plateforme de modélisation spatiale Ocelet (www.ocelet.fr). Le modèle articule plusieurs modules, et permet de simuler des dynamiques de la nappe phréatique, l’étalement urbain, l’avancée du domaine cultivé, les productions agricoles, les revenus agricoles et les emplois agricoles. Le modèle a été construit de manière à reproduire les dynamiques de ces différents modules observés ces 15 dernières années. Il s’agit pour le stagiaire d’explorer la diversité des résultats en entrée et en sortie de modèle et enfin de l’utiliser pour simuler les différents scénarios plausibles.

1 Camara, C., Bourgeois, R., & Jahel, C. (2019). Anticiper l’avenir des territoires agricoles en Afrique de l’Ouest: le cas des Niayes au Sénégal.

Sujet :
La première phase consistera à identifier dans les résultats de sortie du modèle ceux qui correspondent aux scénarios qualitatifs identifiés par les acteurs en 2018 (approche experte). A partir de cet espace des sorties et en utilisant une des méthodes telle qu’OSE, le stagiaire identifiera dans l’espace des entrées les jeux de paramètres qui conduisent aux espaces des sorties considérées par les experts. Pour cela, des séries de simulations seront lancées sur des périodes simulées de 20 ans, en fonction de jeux de paramètres cohérents avec les scénarios qualitatifs produits plus tôt et en insérant différentes « ruptures » dans les simulations (par exemple, introduction d’un nouveau paramètre en cours de simulation). Ce travail d’exploration et d’analyse de l’espace des sorties sera mené par le stagiaire, en s’inspirant là aussi des travaux de la communauté d’OpenMole.
Le stagiaire sera alors à même de produire une interface de visualisation des trajectoires territoriales qui permette aux décideurs et aux chercheurs d’identifier les bifurcations dans les scénarios simulés parmi les avenirs plausibles de la zone des Niayes. Le travail de visualisation des données en sortie – comme par exemple, des cartes d’occurrence de phénomène pour un même scénario, ou une présentation de l’espace des possibles, etc. – fournira le contenu à la plateforme de visualisation.

Profil du candidat :
Durée du stage :
6 mois. Début du stage Avril 2022.

Encadrement :
Le stagiaire sera co-encadré par deux chercheurs du Cirad, Camille Jahel (TETIS) et Etienne Delay (SENS)

Rémunération :
Indemnité de stage en vigueur (environ 573 €/mois).
Prise en charge des frais relatifs aux éventuels déplacements.

Contact :
camille.jahel@cirad.fr
etienne.delay@cirad.fr

Formation et compétences requises :
–

Adresse d’emploi :
Le stagiaire sera accueilli à la maison de la télédétection (www.teledetection.fr), à Montpellier, en fonction du contexte sanitaire en France.

Document attaché : 202202110922_Fouille de modèle et visualisation de données pour explorer les avenirs plausibles de la zone des Niayes au Sénégal_vf.pdf

PhD in machine learning/signal processing

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : CRIStAL (Lille, France), NUS (Singapore)
Durée : 3 years
Contact : remi.bardenet@gmail.com
Date limite de publication : 2022-07-31

Contexte :
URL: http://rbardenet.github.io/pdf/phd-proposal.pdf
Context: “Baccarat” AI chair.
Supervisors: Rémi Bardenet (CNRS, Univ. Lille) and Subhro Ghosh (NUS, Singapore).

A point process is a random discrete set of points in a generic space. A broad interest has emerged in ML and signal processing around point processes that exhibit a regular, repulsive arrangement of their points. For instance, sampling repulsive (i.e., diverse) minibatches yields variance reduction in stochastic gradient descent (Bardenet, Ghosh, and Lin, NeurIPS 2021). As another example, moments of pure silence in the musical score of white noise are a repulsive point process that can be leveraged for signal detection (Bardenet, Flamant, and Chainais, ACHA 2020).

Sujet :
To get acquainted with the interdisciplinary topic of repulsive point processes, we shall start with a project that fits in ongoing collaboration between the two supervisors. Ideally, this project shall be tackled during a master’s level internship prior to starting the PhD. Depending on the student’s background and taste, this can be, e.g., (i) topological data analysis applied to the zeros of random spectrograms (the technical for a time-frequency musical score). Alternately, the internship could revolve around (ii) negatively dependent subsampling for large-scale machine learning. For instance, how can we efficiently build repulsive minibatches in stochastic gradient descent?

After this first project, the three of us will pick an ambitious open problem in line with the objectives of the Baccarat AI chair, according to the student’s interest. Candidate problems include identifying and studying repulsive point processes for high-dimensional Monte Carlo integration, fast sampling algorithms for determinantal point processes in machine learning, dictionary learning for signal processing, or studying zeros of wavelet transforms of random signals to use them in filtering tasks.

Profil du candidat :
The ideal candidate has a strong background in either probability, statistics, ML, or signal processing, and a taste for interdisciplinarity.

Formation et compétences requises :
A master in either probability, statistics, ML, or signal processing.

Adresse d’emploi :
Centre de recherche en informatique, signal et automatique de Lille; Department of Statistics and Data Science, National University of Singapore.

Poste de MCF LIUPPA – IUT de Bayonne et du Pays Basque

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : LIUPPA / T2I
Durée : CDI
Contact : chef-info@iutbayonne.univ-pau.fr
Date limite de publication : 2022-03-31

Contexte :
Un poste de maître de conférences en informatique dans le domaine des architectures logicielles et de l’adaptation au contexte sera ouvert à l’Université de Pau et des Pays de l’Adour (UPPA) au titre de la campagne de concours synchronisée 2022.

Sujet :
L’intégration se fera au sein de l’équipe T2I du LIUPPA (https://liuppa.univ-pau.fr/fr/organisation/equipes-de-recherche/equipe-t2i.html) pour la recherche, et de l’IUT de Bayonne et du Pays Basque (Département Informatique sur le site d’Anglet – Montaury) pour l’enseignement (https://www.iutbayonne.univ-pau.fr).

Profil du candidat :
– Recherche :

Le candidat enseignant-chercheur devra intégrer le LIUPPA au sein de l’équipe T2I, et plus particulièrement dans le domaine de l’adaptation au contexte. Le candidat devra contribuer au niveau des architectures logicielles/middleware, mais il devra également proposer des liens avec les autres thématiques de l’équipe, que ce soit au niveau de l’extraction, de la recherche et/ou l’enrichissement sémantique de données.

Le laboratoire LIUPPA (Laboratoire Informatique de l’Université de Pau et des Pays de l’Adour), équipe d’accueil EA 3000, compte plus de 37 permanents répartis sur trois sites (Pau, Côte Basque et Mont‐de‐Marsan) et dans trois équipes de recherche : Traitements des informations pour l’adaptation de l’interaction au contexte et à l’utilisateur (T2I), Architecture des systèmes cyber-physiques (ASCP), et Génie logiciel (GL). Le LIUPPA positionne son projet scientifique dans un champ applicatif précis : la gestion des systèmes d’information et des architectures des systèmes cyber-physiques (SCP).

– Enseignement et tâches administratives :

Merci de contacter Christophe MARQUESUZAÀ (Chef du Département Informatique de l’IUT de Bayonne et du Pays Basque) : chef-info@iutbayonne.univ-pau.fr

Formation et compétences requises :
Le ou la candidat.e devra être titulaire d’un doctorat en informatique (Section CNU 27).

Adresse d’emploi :
IUT de Bayonne et du Pays Basque – Département Informatique sur le site d’Anglet – Montaury

Dynamic Human-Agent Interactions adapted to users’ profiles

Offre en lien avec l’Action/le Réseau : – — –/Doctorants

Laboratoire/Entreprise : L3i / Easychain
Durée : 36 mois
Contact : mickael.coustaty@univ-lr.fr
Date limite de publication : 2022-04-10

Contexte :
The pandemic that we have just experienced has accelerated the dematerialisation of our exchanges and intensified the need to provide quick and relevant answers at any time. This digital shift practically brings new challenges. Users would maintain the same level of satisfaction in their practices and companies by offering efficient online services. Whether in stores or on the Web, consumers’ expectations are increasing: they want to be able to do everything quickly, easily and efficiently. Therefore, companies should be able to respond to their customers in person (in their establishment) rather than online using their digital platforms. Easychain, a young company that produces software and services dedicated to real estate brokers, aims to anticipate the requirements of its customers and future owners. This problem leads to a strong need to provide answers through chatbots, also known
as conversational agents, in order to satisfy the customers.

EasyChain is a young company based in Niort (France) that publishes tools using Artificial Intelligence. Its first customers are real estate credit brokers, the company addresses the entire real estate sphere, from the buyer to the seller, including notaries, real estate agents, banks and bailiffs. With the arrival of the millenniums on the labour market, the professions connected with real estate will undergo major changes over the next few years: digitalisation and artificial intelligence are the main challenges. In order to meet these future challenges, the company has developed the following tools:
• a tool for recognising and classifying standard and non-standard documents, with extraction of important information, feeding, among other things, a business CRM;
• a mobile application that enables the transfer of spoken and written information, encrypted from end to end. It is a “WhatsApp-like” application extended and dedicated to the real estate credit sphere.

These tools, which are unique on the market, are the very beginning of the automation of the entire credit processing chain. In order to go further in this logic, Easychain aims to complete them with voice chatbots, robots that analyse applications, qualify their eligibility, and process them as a whole. The team is currently composed of AI researchers, mathematicians, software developers, digital specialists,
and business experts.

In order to work on the development of these Chatbot solutions, Easychain would collaborate with the L3i laboratory at La Rochelle University. Created in 1993, the L3i (informatique, image, interaction
computer Science, image, interaction) laboratory is the research laboratory for digital sciences at La Rochelle University. It has about 100 members working in the fields of computer Science, image processing and Interaction. The research fields at L3i concern the interactive and intelligent management of digital content. More specifically, the work carried out within the framework of this collaboration concerns the Images and Contents team. Its core business concerns low-level processing techniques of weakly structured contents (images, texts, videos, native and digitized digital documents, …), as well as the analysis, management and linking of data extracted from such contents (feature extraction, indexing, mining or information retrieval).

In order to improve the quality of its customer relations, Easychain would develop, in collaboration with L3i, a conversational agent. This agent, commonly called Chatbot, should allow:
• answering common questions: customer service is often involved to answer the same questions: problems to login, banking procedures, how to get a home loan, etc. A chatbot will answer these questions. Thus, the support team will have more time devoted to important issues;
• Written or spoken chat: it must be able to consider the customer’s messages either in text or voice and propose an appropriate answer;
• User profile analysis: it is also necessary, not only the emotions, the feelings but also the answer, the way of asking questions, the speech, etc. Based on all the information, the chatbot will be able to adapt its answers.

This chatbot will be integrated in all the tools and services developed by Easychain such as – WeasyFile — a tool for automatic recognition and reading of documents — or ConnectCrédit — a mobile application to ease exchanges between brokers and real estate agencies, etc.

Sujet :
Objectives:
In this particular context, the development of a conversational agent (chatbot) is a necessary response to facilitate communication, by voice or text, with any human being using interactive skills. Generally speaking, chatbots can be seen as computer programs that rely on Machine Learning and Artificial Intelligence (AI) methods to choose the best response based on previous interactions. In this thesis, we want to propose a multimodal conversational agent, able to answer in text or voice according to the modality chosen or entered by the customer. Moreover, this agent will have to be able to extract the domain of the conversation and to detect information about the customer’s feelings (emotion and sentiment analysis) in order to reduce the “robot” effect of existing agents. These objectives are directly linked to two major research areas: Natural Language Processing (NLP) and Natural Language Understanding (NLU). The languages studied by this chatbot will be primarily French and English.

The problem is to take advantage from knowledge and methods dedicated to linguistic analysis such as: morphological analysis, lexical and semantic knowledge, etc.) to achieve two objectives:
– The chatbot must perform a natural language conversation with customers comparable to an exchange between two humans. It is based on language models adapted to the processing of a customer’s file;
– The chatbot must be linked to a voice synthesiser to offer a voice dialogue option with the customer;
– The chatbot must be open domain, meaning that it must be able to adapt its conversation according to its domain;
– The chatbot must be able to analyse the customer’s emotions in real time to determine the best way to respond and dialogue with the customer according to his emotions.

Challenges:
The main challenges are:
– Interpretation of different customer demands: the method must be generic and not based on domain-specific rules;
– Examination of user behaviours: ability to identify and adapt to different customer behaviours, including the modality used to communicate
Brief state of the art and positioning
Several researches have been focused on the design of chatbots. The very first known chatbot was developed in 1966 [1]. It used a simple pattern matching to propose answers (only) to questions. The efforts have then continued on the modelling of a knowledge base used by chatbots such as ontologies, or semantic networks that link a set of hierarchically interconnected concepts [2, 3]. The purpose of using knowledge bases in a chatbot is to compute relations between these concepts, such as synonyms, hyponyms, etc. [3]. The interconnection between these concepts can be represented in a graph allowing the chatbot to search using particular reasoning rules. If the entry is not found in the knowledge base, a default answer is generated. In order to overcome the limits of knowledge bases in terms of coverage of various domains, several research works have therefore used language models in order to semantically analyse the queries (input sentences) with a decent accuracy [4,5].
Existing chatbots suffer from two limitations: 1) model capacity and 2) scarcity of generic data. With the recent success of large pre-trained language models [6-7], which are very effective at encoding semantics [8-9], both problems can be mitigated. The first work to be carried out in this thesis will focus on the use and adaptation of these pre-trained language models on very large databases. The goal consists in generating grammatically and semantically consistent responses rather than adding a domain learning scenario with a goal of classifying dialogue emotions. For example, if the caller says “I am really satisfied with your first article and I look forward the second one”; the chatbot should be able to detect this positive impression of the customer and to adapt in real-time the conversation according to this emotion.
Concerning the opinion analysis and considering the user’s feeling in the answers that the system should give, many works have been done in the literature. The analysis of feelings is a task of Natural Language Processing (NLP). NLP aims at extracting feelings and opinions from texts as presented in [10,11]. In addition, new sentiment analysis techniques are beginning to incorporate information from text and other modalities such as visual data [12,13]. This research topic falls within the field of affective computing and emotion recognition [12]. According to [14], affective computing and sentiment analysis are the keys to the development of artificial intelligence (AI). Moreover, they have great potential when applied to various domains or systems. The task of sentiment analysis can then be viewed as a text classification problem [15-17], as the process involves several operations that result in classifying whether a given text expresses a positive or negative sentiment.
However, although sentiment analysis may seem like an easy process, it actually requires the consideration of many factors not currently addressed by NLP researchers such as sarcasm and subjectivity detection [18,19]. Moreover, the lack of apparent structure (specific to books or newspapers) that can clearly be found in vocal exchanges with clients remains a major problem for the community [20,21].
The proposed method in this research work should define a natural language dialogue system independent of knowledge bases and/or complete models (closed world hypothesis) which are generally cumbersome and incomplete. For this purpose, we propose to study a solution that will rely on attention models [22] to detect the domain and identify its relevant terms. The chatbot will then have to decode the semantics of the terms using recent dynamic word embedding models, which have shown, in our recent work, good performances in several exercises such as information extraction [23] or named entity recognition and disambiguation [24,25]. The major novelty will then consist in developing new models able to integrate information related to the semantics of the content (word embedding, audio embedding [26]), to their context (attention model) and to the user’s feeling (sentiment analysis). The proposed system will then generate a response (text or audio) based on the semantic content, the domain and the user sentiment. The figure below describes a typical architecture of the expected chatbot.

Profil du candidat :
Anybody interested by reasearch in computer science

Formation et compétences requises :
Master degree in computer science (or equivalent)

Adresse d’emploi :

https://l3i.univ-larochelle.fr/spip.php?action=acceder_document&arg=1675&cle=18f8865a07a5837e135038516847ef81c846c59a&file=pdf%2Fcifre_phd_proposal_dynamic_human_agent_interactions_adapted_to_users_profiles_cle07da8a.pdf

Document attaché : 202202100910_cifre_phd_proposal_dynamic_human_agent_interactions_adapted_to_users_profiles_cle07da8a-1.pdf

3 postes EC – Apprentissage/HPC&BDA – Apprentissage et système complexe

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : DVRC/ESILV
Durée : CDI
Contact : nicolas.travers@devinci.fr
Date limite de publication : 2022-03-13

Contexte :
3 postes d’Enseignants-Chercheurs sont ouverts à l’ESILV au laboratoire DVRC.

Sujet :
2 types de profils recherchés liés à l’apprentissage automatique, avec des liens forts avec HPC, BDA ou Systèmes complexes.
Vous trouverez une description sur notre site :
https://www.devinci.fr/carrieres/enseignant-chercheur-informatique-f-h-apprentissage-hpcbda/
et
https://www.devinci.fr/carrieres/enseignant-chercheur-informatique-f-h-algorithmique-et-data/

Profil du candidat :
Le ou la candidat.e devra être titulaire d’un doctorat, une habilitation à diriger des recherches sera appréciée, dans les sections CNU 27 et/ou 61.

Formation et compétences requises :
Le ou la titulaire sera apprécié.e sur :
– La qualité des enseignements dispensés
– La bonne gestion du suivi des étudiants
– Sa coopération avec les services supports en termes de représentation et d’activités transversales
– Sa productivité en recherche scientifique

Adresse d’emploi :
Pôle Léonard de Vinci à Paris La Défense

Poste de professeur en informatique (section 27) au GREYC (Caen)

Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : GREYC UMR 6072
Durée : Permanent
Contact : bruno.cremilleux@unicaen.fr
Date limite de publication : 2022-03-13

Contexte :
Un poste de professeur en informatique (section 27) va être ouvert à l’Université de Caen Normandie au titre de la campagne de recrutement
du printemps 2022.

L’affectation se fera au sein de l’UMR GREYC (dans l’équipe CODAG) et au
département informatique de l’UFR des Sciences.

Sujet :
Les travaux de recherche de CODAG portent notamment sur les thèmes
suivants : les approches déclaratives (SAT, CSP, PL), la fouille de
données, les méthodes quantitatives pour le traitement automatique des
langues, l’ingénierie des connaissances, l’apprentissage et l’aide à la
décision, l’analyse des graphes. L’équipe associe des chercheurs de
cultures scientifiques complémentaires et est particulièrement reconnue
pour ses résultats s’appuyant sur des interactions entre ses thèmes
centraux.

CODAG s’implique fortement dans des projets interdisciplinaires en
menant des collaborations avec, par exemple, des spécialistes du
traitement des données du sport, de l’information chimique ou des
données textuelles…

L’activité de recherche de la personne recrutée portera sur l’un des
thèmes centraux mentionnés, tout en favorisant leurs interactions ou
leurs liens avec les aspects applicatifs. La personne recrutée
renforcera la politique scientifique de l’équipe tout en l’ouvrant vers
de nouveaux défis.

Sur l’aspect enseignement, la personne recrutée sera rattachée au
département mathématiques-informatique de l’UFR des Sciences de
l’université de Caen Normandie et sera susceptible d’enseigner dans
toutes les formations en informatique de ce département, en cycles
licence et master. Des expériences dans les disciplines suivantes seront
appréciées : technologies du web et de l’internet et apprentissage machine.

Profil du candidat :
Mots clé : intelligence artificielle, fouille de données, CSP,
optimisation discrète, théorie des graphes, data analytics, traitement
automatique des langues, web sémantique, protocoles applicatifs

Formation et compétences requises :
Cf. ci-dessus.

Adresse d’emploi :
Contact recherche :
Directeur du laboratoire GREYC :
Christophe Rosenbeger : christophe.rosenberger@ensicaen.fr
Equipe CODAG :
Bertrand Cuissart (responsable) : bertrand.cuissart@unicaen.fr
Patrice Boizumault : patrice.boizumault@unicaen.fr

Contact enseignement :
Directeur du département mathématiques-informatique :
Fabrice Maurel : fabrice.maurel@unicaen.fr