Job description

Détail de l'offre

Informations générales

Entité de rattachement

Le CEA est un acteur majeur de la recherche, au service des citoyens, de l'économie et de l'Etat.

Il apporte des solutions concrètes à leurs besoins dans quatre domaines principaux : transition énergétique, transition numérique, technologies pour la médecine du futur, défense et sécurité sur un socle de recherche fondamentale. Le CEA s'engage depuis plus de 75 ans au service de la souveraineté scientifique, technologique et industrielle de la France et de l'Europe pour un présent et un avenir mieux maîtrisés et plus sûrs.

Implanté au cœur des territoires équipés de très grandes infrastructures de recherche, le CEA dispose d'un large éventail de partenaires académiques et industriels en France, en Europe et à l'international.

Les 20 000 collaboratrices et collaborateurs du CEA partagent trois valeurs fondamentales :

• La conscience des responsabilités
• La coopération
• La curiosité

Référence

2024-32909

Description de l'unité

Located in Saclay, in southern Ile-de-France, CEA LIST is a scientific and technological research center dedicated to the development of software, embedded systems and sensors for applications intended for defense, security, energy, nuclear power, the environment and health. CEA LIST has more than 700 researchers focusing on intelligent digital systems, centered on artificial intelligence, the factory of the future, innovative instrumentation, cyberphysical systems and digital health.

Within this institute, the laboratories of the Digital Instrumentation Department work on the development and industrial transfer of cutting-edge technologies in AI, non-destructive control and nuclear instrumentation. The technical scope of our engineers and researchers concerns the analysis of signals (i.e. time series, but also spectra) produced by equipment developed internally, by CEA teams, or by external companies. The exploitation of this data is based on a broad spectrum of machine learning methods, relating to digital AI (deep neural networks, random forests, SVM) and symbolic AI (knowledge-based systems).

Description du poste

Domaine

Mathématiques, information scientifique, logiciel

Contrat

Post-doctorat

Intitulé de l'offre

Post-doctoral fellowship in AI H/F

Sujet de stage

You will join a team made up of 6 permanent researchers, non-permanent researchers, doctoral students, interns and apprentices, united around the development of an AI called ExpressIF® (https://expressif.cea.fr).

Durée du contrat (en mois)

Description de l'offre

ExpressIF® is a symbolic AI based on knowledge and reasoning, and of which we are also developing the ability to learn knowledge from data, text, images etc. ExpressIF® aims to be an alternative to deep neural networks and other current approaches (random forests, etc.) by offering our partners:
• Solving various problems: decision-making, optimal experimental plan design, planning, etc.
• Trusted AI whose models are understandable by humans and decisions are justifiable.
• An ethical AI which does not aim to replace the human experts but rather to dialogue with them to help them in their tasks.
• An AI capable of learning from little or a lot of data.
As part of a project which concerns the creation of innovative materials, we wish to strengthen our platform in its ability to learn from little experimental data.
In particular, we wish to work initially on the extraction of causal links between manufacturing parameters and property. Causality extraction is a subject of great importance in AI today and we wish to adapt existing approaches to experimental data and their particularities in order to select the variables of interest. Secondly, we will focus on these causal links and their characterization (causal inference) using an approach based on fuzzy rules, that is to say we will create fuzzy rules adapted to their representation.
Bibliography:
[1] Causal discovery for fuzzy rule learning. L Kunitomo-Jacquin, A Lomet, JP Poli. 2022 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)
[2] Towards monotonous functions approximation from few data with Gradual Generalized Modus Ponens: application to materials science. H Hajri, JP Poli, L Boudet. 2021 IEEE 33rd International Conference on Tools with Artificial Intelligence
[3] Towards an interpretable fuzzy approach to experimental design. O. Rousselle, J-P Poli, N. Ben Abdallah. IPMU 2024.

#CEA-List ; #PhD ; #Artificial Intelligence ; #AI

Desired profile

Profil du candidat

The candidate will have a doctorate in AI, statistics, applied mathematics or computer science.

The ability to work in a team is necessary, while demonstrating autonomy in daily tasks. As developments in AI advance rapidly, it is necessary for the candidate to have the ability to continually renew and enrich their skills. Finally, excellent expression skills, both oral and written, in English, will be necessary for you to communicate with our many partners. A taste for applied research is also important. Prototyping in python will be requested.

Offers “CEA”