THESIS OFFER FINANCED 2022 - SYSREIC / LISIC / ULCO

TEAM Title of the thesis project: Automatic Classifiers of Ontologically Explainable Images: application to the Observation and Analysis of Marine Flying Fauna.

Keywords: Artificial Intelligence, Computer Vision, Knowledge Engineering, Explainability of AI

Funding: Hauts de France Region / ULCO (Thesis scholarship)

Contact: [email protected], mourad.bouneffa@univ-littoral. En

Summary : Responding to the need for explainability of AIs that use Deep Learning, we propose in this thesis to develop a hybrid approach in Computer Vision for the automatic classification of images combining machine learning models with semantic knowledge expressed by ontologies, with the aim of creating ontologically explainable image classifiers. The SysReIC team has already achieved very encouraging results in a prototype. This thesis will consist in developing these ideas by applying them to the ecological problem of monitoring and observing the trajectories of flying fauna in partnership with the companies Écosphère (1, Antenne Nord Littoral) and Prodrone (2). This type of study is often a prerequisite for impact analysis for the installation of infrastructure both at sea (offshore platforms, ...) and on land (industrial installations, bridges, ...). This process is now carried out by overflight campaigns of sites studied by ornithologists in aircraft. This approach is costly, dangerous, and moderately effective. To overcome these problems, it is proposed to implement a new process using autonomous flying image capture devices (wings and drones), and to carry out image analysis using AI techniques. In this context, the importance of the issues requires the implementation of the expert knowledge necessary to refine the analysis process, but also to provide results that are explainable to specialists.

(1) https://www.ecosphere.fr/

(2) https://www.prodrones.fr/

Coming up with plans to achieve their goals---and more generally deciding and learning how to act---are fundamental problems for intelligent agents. The traditional focus of AI planning is the generation of plans for an individual agent under the hypothesis that no other agent intervenes and that the planning agent has perfect knowledge of the environment and action pre- and postconditions. More expressive forms of planning should include things such as reasoning about other agents' beliefs, goals, intentions and actions ('theory of mind');

learning from past experiences; monitoring the actual outcome of actions and learning possibly unexpected outcomes and dealing with such new outcomes. Formal tools such as logics of knowledge and belief, logics of goals, intentions and actions, game theory, and learning can be expected to be building blocks of such formalisms.

Applicants should propose a short research project related to these themes, possibly relating it to previous work of ours where we have contributed to epistemic planning and to planning inspired from Bratman's theory of intention.

The duration of the contract is 18 months (possibly with an extension). Gross salary is between 2.663€ (about 2.131€ net) and 3.783€ (about 3.027€ net), depending on the experience of the candidate. Starting date is September 1, 2022.

* Context The Institut de Recherche en Informatique de Toulouse (IRIT) is one of the biggest computer science labs in France. Its AI Department has a long-standing tradition of research in knowledge representation and reasoning. IRIT participates in the ANITI project on hybrid AI.

The postdoc position is within the TAILOR network (Trustworthy AI: Integrating Learning, Optimization and Reasoning, https://tailor-network.eu/) in which CNRS-IRIT takes part, and is part of its workpackage WP5 "Deciding and Learning how to Act", possibly in connection with WP6 "Learning and reasoning in social contexts". TAILOR is one of four networks of excellence working on aspects of trustworthy AI funded under the H2020-ICT-48-2020 call.

* Profile and Application We are looking for a candidate with a background in knowledge representation and/or machine learning who is interested in their integration in the context of planning. Candidates should send the following to [email protected] before April 30:

• CV;

• Up to 2 reference letters;

• A short proposal of research activities (max 1 page).

The Laboratory of Computer Science and Digital Society (LIST3N) at UTT is recruiting a doctoral student for a doctoral thesis in Computer Science. The subject is attached.

Interested candidates, please send us your application before May 10, 2022.

Desired start on 01/10/2022.

Application deadline: May 10, 2022

Duration: 3 years from October 2022.

Candidate Profiles

• Hold a research master's degree in Computer Science, Applied Mathematics or a diploma engineer with research activities.

• Advanced programming capacity (in C, C++, python).

• Very good knowledge of combinatorial optimization tools (linear programming, meta- heuristics).

• Knowledge of AI and machine learning is a plus.

• Experience in developing resolution methods for routing problems of vehicles will be appreciated.

• Good level in English.

Application File: Please send your application in the form of a pdf file (if possible a single file) including a detailed curriculum vitae, transcripts, copies of the most recent master's or engineering diplomas as well as, as far as possible, a cover letter and any recommendations and/or the internship report to the contact emails below:

Contact : • Oumayma BAHRI ([email protected]), LIST3N Lecturer, UTT.

• Lionel Amodeo [email protected]), University Professor, LIST3N, UTT.

The Lattice laboratory, in collaboration with sciencesPo's medialab, is offering a doctoral contract in computational social sciences, on the following theme: "Computational study of the circulation of statements in the digital public space" Co-direction between Thierry Poibeau (Lattice -- CNRS & ENS/PSL & U. Sorbonne nouvelle) and Sylvain Parasie (Medialab SciencesPo)

Full description: https://euraxess.ec.europa.eu/ jobs/763078

Application deadline: April 15, 2022 (all information

to apply is available from the link above)

The thesis will be the subject of a 3-year doctoral contract within the framework of the ANR Medialex

*** Abstract *** The thesis is part of the MEDIALEX project, funded by the ANR (2021-2025). This interdisciplinary project, which brings together sociologists, political scientists, computer scientists, linguists and economists with a common interest in computational methods, aims to determine the logics of formation of the political and media agenda.

The thesis will focus on the large-scale study of the circulation of statements (such as quotations or small sentences pronounced by public figures ) in the public space (audiovisual, print and web media; social networks; parliament). It will be a question of developing NLP techniques allowing the identification on a large scale both of the circulation/distortion of the statements, but also the identification of the speakers and their ideological positioning. We will start from the emblematic (but already old) works of the field, in particular those carried out by the teams of Lada Adamic or Jure Leskovec for example (see references below). It will be necessary to see how recent techniques of automatic language processing can complement the analyses proposed in the past, by allowing a better consideration of the context, or by trying to move towards an automatic (unsupervised) identification of quotations and "small sentences" type.

The data on which the thesis will focus will come from Twitter, Facebook, the main French audiovisual media (from the National Audiovisual Institute, partner of the project), web media , in particular. These corpora will be partly open.

In collaboration with the Medialex project team, the thesis will seek to contribute to contemporary discussions on the structure and dynamics of the public sphere in the digital age (Cardon, 2010).

More details: https://euraxess.ec.europa.eu/ jobs/763078

A postdoctoral position is available in the SPARKS team at the I3S laboratory, Université Côte d’Azur, France (see https://www.i3s.unice.fr/ for more information on the lab).

The bioinformatics group focuses on using network approaches to analyze and integrate large-scale ‘omic data, and on developing computational tools to model how perturbations in gene regulation can affect biological processes.

Within the framework of a project whose objective is to develop a new test capable of detecting and quantifying the risk linked to non-genotoxic carcinogenic substances (NGTxC), we plan to take a holistic approach to better understand the processes at work at the scale of the biological system. By integrating public data and genomic data generated by our biologist partners (including coding but also non-coding genes), we will model the data as networks. In systems biology, a network maps molecular entities (e.g. genes, transcripts, proteins) via their functional interconnections (which can be physical interactions, transcriptional inductions, enzymatic activation, etc). This modeling, taking into account the individual components and their interactions, will allow us to identify subparts of the networks that are particularly active in NGTxC-induced deregulations. Efficient algorithms to discover these active modules in complex networks have been proposed [1] and are being studied by the SPARKS team [2-3].

The mission ----------- The aim is to extend current network analysis methods along several axes by (1) enabling the processing of data obtained from Single-cell RNA sequencing and Long-Read RNA Sequencing, (2) integrating other data into the analysis pipeline, wether new types of high-throughput biological data or other knowledge represented as networks, (3) taking into account the temporal aspect of the data by including measurements made at several time points.

For this, the candidate will have to rely on recent developments concerning community detection in temporal networks [4], community detection in attributed networks [5], multi-view clustering [6] and of course machine learning applied to single cell data analysis [7].

Key responsabilities -------------------- The person recruited will be in charge of the following:

- Literature review of software solutions to apprehend the underlying strategies for extracting knowledge from complex network structures,

- Practical tests to evaluate the efficiency and ease of use of selected tools in the context of high-throughput analyses,

- Determination and implementation of optimal strategies for data extraction,

- analysis of the findings in collaboration with biologists.

Profile ------- - PhD in computational biology, biostatistics or computer science

- Experience with analysis of high-throughput ‘omics data

- Proficiency in programming (experience with Python and its ecosystem is preferred)

- Knowledge in single cell transcriptomics, gene regulation and network biology is desirable

- Experience with high performance computing is a plus

- Ability to think and work independently, set goals and meet deadlines

- Professional proficiency in English

- Good communication and writing skills

- Willingness to work in a multidisciplinary environment, sharing skills and ideas

Work environment ---------------- Located in the Sophia Antipolis technology park, between Nice and Cannes, the I3S laboratory (Computer Science, Signals and Systems of Sophia Antipolis - CNRS UMR 7271) employs 230 people, including about 100 researchers and professors and about 80 PhD students. The SPARKS team (Scalable and Pervasive softwARe and Knowledge Systems) is the largest team at I3S with a staff of 104, including 44 permanent staff.

The team studies the organization, representation and distributed processing of knowledge, as well as its extraction from data and its semantic formalization, with a particular focus on scaling up and designing adaptive knowledge-centric and human-centric software systems. The team is structured around four themes: "knowledge extraction and learning" (machine learning, data mining, artificial intelligence), "formalization and reasoning between users and models" (multidisciplinary approaches to multi-criteria analysis and modeling, reasoning about these models using graph-oriented approaches of the Semantic Web), "scalable software systems" (adapting and composing systems, data and workflows at different scales, from local loop to massive distribution), "computer science and biology" (knowledge extraction, modeling and simulation of dynamic biological systems, formal proofs of the behavior of biological systems and computer-aided model-based reasoning).

The work will be carried out within the framework of the NewgenTOXiv project financed by the 4th Future Investment Programme (AIP 4) of the French government. The project involves 3 public research laboratories (I3S, IPMC and ICN) and two industrial companies (ImmunoSearch and MyDataModels).

Type of contract : fixed-term contract, full-time position Contract duration: 24 months

Desired hiring date : 1st june 2022

Place of work: I3S laboratory in Sophia Antipolis

Gross salary: from 2500€ to 3800€, depending on experience

Applications should include the following documents in electronic format: - a cover letter, stating your motivation, scientific background, and research interests,

- a detailed CV with a list of publications,

- 2-3 references (name, institution, e-mail, telephone number, and relation to the candidate).

Send all these documents by email to [email protected]

A pdf version of the position is available at https://claude.pasquier.net/data/postdoc_2022.pdf

We are proposing a thesis in collaboration with Orange on the theme of the digital twin.

For more information and to apply: https://orange.jobs/jobs/offer.do?joid=111950&lang=EN

The IRT SystemX (Saclay plateau) offers an engineer-researcher (M/F) position in the field of knowledge management, ontologies, semantic web and learning.

For more information, consult the offer: https://www.irt-systemx.fr/ recruitment/ingenieur-chercheur-en-gestion-de-connaissances-ontologies-web-semantique-et-apprentissage-f-h/

Proposition de projet de recherche doctoral Campagne SCAI 2022

Supervision Marie-Jeanne Lesot, LIP6, LFI team, axis: Artificial intelligence and data science Carola Doerr, LIP6, RO team, axis: Theory and mathematics of computing

Affiliation Lab: LIP6, UMR7606, SU, CNRS

Ecole Doctorale: ED130 - EDITE

Start date: Octobre 2022

Keywords: ontology, conceptual graphs, graph pattern mining, fusion and aggregation, black-box

optimisation, algorithm selection, benchmarking, XAI

Candidate profile: A Master’s degree in a quantitative field such as Computer Science, Engineering,

Statistics, Operations Research, Mathematics is required. We expect willingness to conduct empiri- cal research as well as experience with the python programming language. Since the student will be working in an international research team, they must be proficient in written and spoken English.

Knowledge of French is not required. International students are very welcome to apply.

How to apply? Deadline: April 26th

https://www.sorbonne-universite.fr/projets-proposes-en-2022-programme-instituts-et-initiatives In addition, send a cv, motivation letter, grades obtained in master, recommendation letters when possible to the supervisors, Marie-Jeanne Lesot and Carola Doerr

The general aim of the thesis is to exploit expert knowledge regarding properties of optimisation al- gorithms and problems, represented in the formal frameworks of ontologies and conceptual graphs, and to develop tools to extract automatically underlying correlations: the objective is to allow under- standing the reasons why an algorithm is more appropriate than others to solve a problem depending on its characterisation and possibly to offer new tools to configure optimisation algorithms.

To do so, the project will explore new methods for analysing conceptual graphs and in particular design dedicated frequent pattern mining algorithms. The output of such algorithms is also represented in the framework of conceptual graphs, which makes the results more legible and understandable for the end user.

The thesis is expected to contribute at the cross-roads of the domains of knowledge representation, pattern mining and black-box optimisation.

We offer a thesis offer within the "Interactive Computing" team of the ENAC laboratory.

Please distribute this offer to potentially interested students.

Title: Formal Verification of Interactive Systems by Deductive Approach

Keywords : Reactive software, Formal deductive verification, Lowest precondition

The objective of this thesis is to contribute to the formal verification of graphic properties on reactive languages by developing a deductive approach based on the calculation of weaker preconditions.

This objective is original because on the one hand the approach aims to adapt to reactive languages a formal verification approach that is classic in the field of imperative languages and on the other hand it seeks to formally verify properties relating to the human-machine interface, particularly graphic properties.

Location: National School of Civil Aviation

Toulouse – France

Dates : Applications expected until 30/04/2022

Thesis is scheduled to start: October 2022

Duration: 3 years

Details can be found in the attached pdf file or here: https://www.enac.fr/fr/ job offer

Hello everyone, The DUKe (Data User Knowledge) team of LS2N (Laboratoire des sciences du numérique de Nantes), UMR CNRS 6004 (https://www.ls2n.fr) and chekk (https://www.chekk.me) are launching a call for applications for a CIFRE doctoral position in the field of knowledge graphs and machine learning.

Title: Integration of semantic knowledge into a graph diving approach for improving the quality of knowledge graphs

Keywords: Entity resolution, Knowledge graph, Machine learning, Deep learning, Graph diving, Ontology.

Please find attached a full description of this offer. Do not hesitate to spread it around you.

Mounira Harzallah Maître de Conférences HC HdR - Associate professor https://pagespersowp.ls2n.fr/mouniraharzallah/

Halmstad University, School of Information Technology “Our School and research environment is in a very exciting and expansive phase. It is a large research environment with a strong international character. The goal is to be a nationally leading environment within, among other things, applied AI” – Magnus Clarin, Dean of the School of Information Technology.

Halmstad University Halmstad University adds value, drives innovation, and prepares people and society for the future. Since the beginning in 1983, the University has been characterised as forward-thinking and collaborative. Halmstad University offers popular and reality-based education programmes. The research is profiled within two focus areas: Health Innovation and Smart Cities and Communities. The research at Halmstad University is internationally renowned and is pursued in multidisciplinary innovation and research programs. The University takes an active part in the development of society through extensive and recognised collaboration with both the private and public sector.

More information about working at Halmstad University: https://hh.se/english/about-the-university/vacant-positions.html

The School of Information Technology Halmstad University consists of four interdisciplinary Schools and the current position is located at the School of Information Technology (ITE). ITE is a multicultural school with around 130 employees from 20 different countries. It is a strong research and education environment, with focus on smart technology and its applications. Students and researchers are working with everything from AI and information-driven care to autonomous vehicles, social robotics, and digital design. ITE offers education on all levels, from undergraduate to PhD education, plus education for professional. Research is conducted within aware intelligent systems, smart electronic systems, cyber physical systems and digital service innovation. These four areas constitute the four technology areas of ITE. An innovation centre for information-driven care called Leap for Life is connected to ITE, as well as a collaboration arena for electronic development, Electronics Centre in Halmstad (ECH).

More information about the School of Information Technology: hh.se/ite-en

Description The selected PhD students will carry out research on developing AI-based novel sensor denoising and adaptive fusion approaches to have robust in-vehicle perception algorithms for autonomous vehicles. More specifically, the PhD works involve the following topics:

Developing a novel generative deep neural network model to detect, identify, and filter out noisy Camera, LiDAR, and RADAR readings affected by rain, fog, snow, and slush.

Developing a novel adaptive sensor fusion algorithm based on optimum cost-effective sensor combinations to handle possible sensor failures in a multi sensor setup.

Both PhD positions are offered in the frame of a Horizon Europe project ROADVIEW (Robust Automated Driving in Extreme Weather) coordinated by Prof. Aksoy from Halmstad University.

The selected PhD students will be responsible for conducting research within the ROADVIEW project and for participating in the required PhD course activities. The employments also includes teaching responsibilities corresponding to a maximum of 20% of full-time. Applicants will be part of an international workgroup and an English-speaking environment.

This is two full-time positions available from September 2022 for a period of four years. Enough resources to fund experiments and conference travels are available.

Work environment The positions is at CAISR (Center for Applied Intelligent Systems Research), which is a long-term research program on intelligent systems at the School of Information Technology at Halmstad University. The focus of the centre is on aware intelligent systems, computer systems that are human aware, situation aware and to some extent “self-aware". This means systems that can communicate with humans, systems that can be aware of human intentions, systems that can fuse information and be aware of the situation and not just individual sensor readings, and systems that can be self-monitoring.

Qualifications The ideal candidate has a Master’s degree in computer science, machine learning, robotics, mathematics, or a related engineering discipline. A strong background in computer vision, machine learning, artificial intelligence, data mining, or signal processing is desirable. Excellent programming skills, analytical problem solving and organizational abilities are required. Prior practical experience in deep learning is a plus.

Students expecting to finalize their degree in the coming months are also welcome to apply. Only those who are or have been admitted to third-cycle courses and study programmes at a higher education may be appointed to doctoral studentships. (The Higher Education Ordinance Chapter 5 Section 3). The student’s ability to benefit from doctoral studies will be taken into account when we make the appointment. (The Higher Education Ordinance Chapter 5 Section 5).

Salary Doctoral students are employees of the University and paid a salary according to a uniform salary scale, adjusted in relation to the progress in education.

The position includes studies up to a doctoral degree and is annually extended in accordance with the The Higher Education Ordinance Chapter 5 Section 7. The total employment period is four years but may be extended to a maximum of five years if the student performs 20% teaching or other tasks within the university.

Applications should be sent via Halmstad University's recruitment system Varbi (see link here), and should include the following documents.

1. A cover letter stating the purpose of the application and a brief statement of why you believe that your background and goals are well-matched with the goals of this position

2. A Curriculum Vitae that includes at least: a list of previous degrees, dates, and institution, transcripts for higher-education studies until most recent available

3. Copies of previous transcripts and degree certificates

4. A summary (1-2 pages) of the master’s thesis

5. A copy of previous publications and software samples, if any, and

6. Two reference letters and contact information for three reference persons

List of qualifications and other documents that the applicant wishes to refer to should be enclosed with the application. All copies must be attested.

General Information We value the qualities that gender balance and diversity bring to our organization. We therefore welcome applicants with different backgrounds, gender, functionality and, not least, life experience.

Read more about Halmstad University at http://hh.se/english/discover/discoverhalmstaduniversity. 9285.html

Apply Now https://hh.varbi.com/se/what:job/jobID:478735/iframeEmbedded:0/where:4

Type of employment Temporary position longer than 6 months

Contract type Full time

First day of employment 2022-09-01 or as soon as possible

Salary Monthly salary

Number of positions 2

Working hours 100%

City Halmstad

County Halland county

Country Sweden

Reference number 2022/23

Contact Håkan Pettersson, +46 35 167306 Stefan Byttner, +46729773601

Published 31.Mar.2022

Last application date 20.Apr.2022 11:59 PM CEST

Postdoc in Computer Vision / Machine Learning / Applied Mathematics

The Division of Computational Science and Technology at KTH Royal Institute of Technology in Stockholm, Sweden is seeking a Postdoc in Computer Vision / Machine Learning / Applied Mathematics to handle scale-dependent image information in deep networks.

In our research, we develop deep networks for processing image data that handle scaling transformations and other image transformations in a theoretically well-founded manner. Our research in this area comprises both theoretical modelling of the influence of image transformations on different architectures for deep networks and the experimental evaluation of such networks on benchmark datasets to explore their properties. The work also comprises the creation of new benchmark datasets, to enable characterization of properties of deep networks that are not covered by existing datasets. For examples of our previous work in this area, see https://www.kth.se/profile/tony/page/deep-networks

Within the scope of this postdoc position, you are expected to work on and contribute to the research frontier regarding scale-covariant or scale-equivariant deep networks and/or deep networks parameterized in terms of Gaussian derivatives, on specific research topics that we choose together.

The selected candidate will work closely together with the project leader Tony Lindeberg.

A doctoral degree or an equivalent foreign degree, obtained within the last three years prior to the application deadline

We are seeking a candidate with a PhD in Computer Vision, Machine Learning, Applied Mathematics or a related discipline dealing with automated analysis of image information.

Previous experience with experimental evaluations using deep learning architectures applied to image data is necessary, preferably PyTorch.

A theoretical background in continuous mathematics for modelling convolutions and the influence of image transformations on image data is also necessary.

As a person you have excellent scientific and collaborative skills, in combination with independence, with very good ability to get into new scientific theories and conduct implementations and experimental evaluations in close collaboration with the research environment you are working in.

The preferred candidates should have demonstrated expertise (through publications) in any one of the following:

Deep networks that handle image information for computer vision tasks, including experimental evaluation using modern architectures for deep networks.

Continuous models for deep networks applied to image information.

Theoretical modelling of scaling transformations or other image transformations applied to automated processing of image information.

For further information and information about to how to apply, see https://www.kth.se/en/om/work-at-kth/lediga-jobb/what:job/jobID:490734/

Application deadline: May 2, 2022

The position is offered for a period of two years.

Full-time PhD student in Process Algebras for Cybersecurity (CyberExcellence project)

University faculty : Computer science

Grade : researcher

Contract : renewable fixed term contract

Category : scientific personnel

Allocation : External funds

Reference : External funds (CyberExcellence project).

The University of Namur (UNamur) is located in the centre of Belgium, in the French-speaking part of the country. It offers quality education to more than 7,000 students every year and hosts more than 900 researchers in all fields of expertise. The Faculty of Computer Science provides cutting-edge teaching and research, intending to put computers at the service of society, taking account of their impact on the environment and respecting the values of solidarity and sustainable development. The Faculty is a founding member of the Namur Digital Institute (NADI), which groups together over 150 researchers in digital technology. It has a multi-disciplinary approach and addresses, in particular, the issues and challenges of computer science in organisations and society. The Faculty of Computer Science has over 400 students, 80 members of staff, including 18 professors and around 50 researchers. Founded in 1968, the Faculty has trained over 1,800 high-level computer science graduates since then.

The CyberExcellence project started in January 2022. It aims at positioning Wallonia as a major player in cybersecurity on the national and international map by developing a core framework allowing the implementation of solutions based on practical and thoughtful cybersecurity with a competitive advantage.

Among the different axes addressed by the project, axis 1 aims to secure the system before its deployment through formal approaches. In the context of the proposed position, we plan to use process algebras based on the Bach coordination language developed at the University of Namur to model systems and analyze them to detect vulnerabilities. In a complementary approach, we also plan to enrich coordination languages with security mechanisms.

More information: https://www.digitalwallonia. be/fr/publications/cyberexcellence-projet-recherche-cybersecurite.

The position is a 2-years (renewable) contract, funded by the CyberExcellence project. The person hired will join the Focus research center (https://directory.unamur.be/entities/focus) under the supervision of Prof. J.-M. Jacquet, I. Linden and J.-N. Colin. He will be actively involved in research leading to a doctoral thesis within the Namur Digital Institute (https://nadi.unamur.be). This thesis will aim at the design and implementation of coordination-based process algebra to model and analyze systems. This research will be conducted in collaboration with the various academic and industrial partners of the project.

The objectives of the thesis are the following:

Study current process algebras approaches to model systems and detect vulnerabilities

Extend the Bach language with security mechanisms and implement these extensions

Study the algebraic semantics of this language (and more generally process algebras based on asynchronous communication)

Apply these semantics to detect security issues

Those who apply will carry:

a Master's degree (120 ECTS credits) in Computer Science, or equivalent, or a Civil Engineering degree with good computer skills, or equivalent.

The person hired will demonstrate:

interest for formal methods and concurrency theory;

ability to integrate and work in a research team;

good communication and presentation skills in English (read, written, spoken), knowledge of French is considered as a plus,

sense of initiative and responsibility, autonomy, and organization.

Note that soon to be graduating master students are welcome to apply provided that they will have graduated before the start of the position. Candidates that already hold a PhD degree are not eligible.

Additional information For additional information please do not hesitate to contact:

J.-M. Jacquet - contact: [email protected] I. Linden - contact: [email protected]

Remarks Contract: fixed term contract, two-years (renewable).

Expected starting date: September 1st, 2022 (negotiable, depending on COVID or candidate constraints).

How to apply Applications should be sent by e-mail to [email protected] AND [email protected] and [email protected]. They should contain the following:

a motivation letter describing the interest in the research topic; a recent CV; a copy of diplomas (Bachelor and Master, if available); the name and e-mail address of one reference person to be contacted upon request.

A dedicated selection committee will examine applications.

Submission deadline: May, 31th, 2022 (11h59 pm AoE).

We are looking for a candidate for a thesis subject in Artificial Intelligence / Machine Learning, for an application in Health.

Start date September or October 2022

Topic The thesis work is part of a highly interdisciplinary context (AI / Health). The aim of the thesis is to develop an automated manager of anesthesia simulation scenarios on a digital patient, for the training of interns and nurse anesthetists. These scenarios must be reactive to the actions of the intern or the nurse-in-training. The key to the problem consists in knowing how to simulate the evolution of the physiological parameters of the digital patient according to the action triggered by the person in training or by the rest of the medical team (which is virtual). It is also necessary to know how to predict the next action of the medical team (excluding the person in training). We can take advantage of the data recorded during the anesthesia of a cohort of real patients, operated for the same surgery, to predict the evolution of the digital patient over the actions performed on the patient. For each patient in the cohort, we also have a record of the medical actions performed on the patient, in the operating room. These traces are used to predict the next action to perform, over a scenario. From an AI point of view, we are therefore faced with a problem of semi-supervised learning from multivariate time series and interdependent event traces, with the aim of predicting short-term and real-time multivariate time series. , and prediction of the next event.

Management Christine Sinoquet (web page), Thesis director, HdR Lecturer in Computer Science at the Nantes Digital Sciences Laboratory / UMR CNRS 6004 and Corinne Lejus-Bourdeau, University Professor – Hospital Practitioner, Doctor of Medicine in the Anesthesia-Surgical Resuscitation Department of the University Hospital of Nantes / Hôtel Dieu – Women-Child-Adolescent Hospital, Director of the Experimental Laboratory for Simulation of Intensive Medicine of the University of Nantes (LESiMU).

Keywords Artificial intelligence, health of the future, digital patient, digitally assisted training, operating room, anesthesia, simulation, modeling, semi-supervised learning, deep learning, time series, traces of events

Profile Master's degree or equivalent in Mathematics or Mathematics / Computer Science or Computer Science

with specialization in data science or probability / statistics, as well as in machine learning (including deep learning preferably)

- Theoretical skills and experience required in probability/statistics, applied mathematics, machine learning

- If the candidate does not have experience in modeling by composition of models using deep neural networks, which is one of the axes of investigation of the thesis, he must on the other hand show a strong motivation to s invest in this area

- Interest in interdisciplinarity (health)

- Programming experience and good level of programming

- Good writing skills

- Ability to work in a team, ability to report on the progress of its work

Funding This research will be funded under the AIby4 (AIby4) project. AIby4 is one of the 22 projects selected by the ANR for its call for “doctoral contracts in AI” (2020-25).

Selection calendar Applications will preferably be sent before Thursday, April 21, 2022 12:00 p.m.

Deadline for sending applications: Tuesday, April 26, 2022 12 noon

People corresponding to the requested profile will be called for an audition by videoconference (over the water).

Decision: mid-May 2022

Documents required - Detailed CV

- cover letter

- Master 1 transcript (with ranking rank and class size)

- Master 2 notes excluding internship (with ranking rank and class size)

- summary of the current internship (between 2 and 4 pages, bibliographical references not included)

- letters of recommendation for the current year

- contact details of reference persons (first name, last name, status, institution (detail the acronyms if applicable), city, email address, telephone number)

Questions and sending application files (zip archive) to [email protected]