Post doc

Organisation/Company CNRS Department Laboratoire d'informatique de modélisation et d'optimisation des systèmes Research Field Mathematics History » History of science Researcher Profile First Stage Researcher (R1) Application Deadline 20 Oct 2026 - 23:59 (UTC) Country France Type of Contract Temporary Job Status Full-time Hours Per Week 35 Offer Starting Date 29 Oct 2026 Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Is the Job related to staff position within a Research Infrastructure? No

Offer Description

The postdoctoral researcher will develop novel methods based on probabilistic logic and graphical models for reasoning and query answering over incomplete data.

The research will focus in particular on exploiting a priori knowledge, represented through causal graphs, Bayesian networks, or Markov networks, to model dependencies between variables and to define probability distributions over possible completions of incomplete databases.

The researcher will contribute to the development of formal models, inference and query-answering algorithms, and their experimental validation.

Develop probabilistic and logical models for representing incomplete data and missingness mechanisms.
Exploit directed causal graphs and undirected Markov networks as a priori knowledge about dependencies between variables.
Define probabilistic semantics for possible completions of incomplete databases.
Investigate the integration of logical constraints, rules, and domain knowledge into probabilistic models.
Develop methods for probabilistic inference and query answering over incomplete data.
Study the theoretical properties and computational complexity of the proposed inference and query-answering problems.
Design and implement research prototypes and evaluate the proposed approaches on synthetic and real-world datasets.
Publish research results in international conferences and journals.
Participate in project meetings and contribute to the dissemination and valorization of research results.

The position will be hosted by the Laboratory of Computer Science, Modelling and Optimization of Systems (LIMOS, UMR CNRS 6158), a joint research unit in computer science under the supervision of the French National Centre for Scientific Research (CNRS), Université Clermont Auvergne (UCA), and École des Mines de Saint-Étienne (EMSE), with Clermont Auvergne INP as a secondary supervisory institution. LIMOS is affiliated with the CNRS Institute for Computer Science.

The laboratory is primarily located on the Clermont-Ferrand/Aubière sites, on the Cézeaux campus, and in Saint-Étienne. The position will be based on the Cézeaux campus in Aubière, within the scientific ecosystem of Université Clermont Auvergne and Clermont Auvergne INP.

LIMOS brings together a large scientific community, currently comprising more than 200 members, including approximately 96 faculty members and researchers, 70 PhD students, and around 15 postdoctoral researchers, as well as research support, administrative, and technical staff.

The laboratory's scientific activities are structured around three main research areas: MAAD — Models and Algorithms for Decision Support, SIC — Information and Communication Systems, and ODPS — Decision-Making Tools for Production and Services.

The postdoctoral researcher will join the SIC research area, whose work focuses in particular on data acquisition, management, and analysis, as well as on issues related to data quality, interoperability, and information processing.

Within SIC, the researcher will work in the DSI — Data, Services and Intelligence research theme, which brings together research on the management and optimization of large-scale data, the integration and querying of heterogeneous data, as well as knowledge extraction and machine learning. The theme also addresses the processing of complex data and the scalability of artificial intelligence methods.

Good knowledge of artificial intelligence and machine learning.
Knowledge of logic, knowledge representation, or automated reasoning.
Knowledge of databases, incomplete data, or probabilistic databases.
Knowledge of probabilistic graphical models, such as Bayesian networks, Markov networks, and/or causal models.
Experience with probabilistic logic programming, probabilistic reasoning, or neuro-symbolic approaches would be an advantage.