Post-Doc in "Scientific Machine Learning and Mechanics for scale-bridging in granular materials"

Contexte et atouts du poste The appointment is part of the Chair "Artificial Intelligence and Mechanics for scale bridging in complex materials" (AIM), funded by MIAI Cluster AI and the ANR through the France 2030 program. The research will be conducted at Inria and UGA (Grenoble, France). The position provides the opportunity to work on a challenging, high-impact research topic at the crossroad of AI, physics, and mechanics, with excellent prospects for careers in both academia and industry. Granular materials exhibit complex and distinctive behaviours arising from the disordered spatial arrangement of microscopic solid particles (grains). These materials are central to a wide spectrum of industrial and environmental processes. Yet, the mechanics of granular systems remain opaque due to their multiscale, nonlinear, and history-dependent behaviour. At the microscale, grain interactions—governed by nonsmooth phenomena, e.g. contacts, friction—drive the bulk dynamics at the macroscale, which challenge current modelling frameworks. Modern experimental methods provide unprecedented quantitative observations, but they still cannot fully access the internal material state (e.g. contact forces, grain deformation). High-fidelity particle simulations represent reliable probes of the mechanics at the microscale but remain too computationally intensive at larger scales. Macroscopic, continuum-based models are hence often preferred for their computational efficiency. However, these models use heuristic constitutive equations that require phenomenological parameterisations based on conventional laboratory tests, not accounting for the rich measurement data that can be extracted today with in-operando experiments (e.g. tomography). AIM's interdisciplinary methodology bridges applied mathematics, mechanics, and artificial intelligence to better understand, model, and predict the mechanics and dynamics of granular media. Based on high-fidelity particle-scale simulations and cutting-edge in-operando experiments, data-driven methods will be developed to discover physics-based material descriptors and build a novel multiscale approach to robustly and accurately predict the fine- and large-scale behaviour of granular systems. Mission confiée The appointment is for a duration of two years. The successful candidate will be appointed at UGA and Inria and will join the AIM research group, with members from TRIPOP, THOTH project teams (Inria), the Geomechanics group at 3SR, and the ECRINS team (INRAE-IGE). The candidate will be based at the Inria Center of Grenoble Alpes University and UGA, including an engaging and collaborative research environment with access to state-of-the-art computing resources, field and laboratory facilities, and numerous opportunities for professional development and collaboration. The position also includes opportunities to engage in academic activities, such as supervising Master’s and undergraduate students. The project includes funding for travel to international conferences and research visits, fostering the dissemination of the findings and collaborations within the academic community. Principales activités The research topic focuses on fundamental developments of a novel learning framework for the discovery of robust and accurate behaviour equations of granular media. While unknown, these laws must obey general principles from statistical mechanics and thermodynamics, which reduces the search space and opens the way for uncovering them using data-driven techniques and bridge the gap between constitutive modelling, numerical simulations, and experiments. As a Postdoctoral Researcher, you will explore and develop AI methods to hardwire physical principles and discover interpretable, physics-based statistical descriptors of the microstructure from high-fidelity simulations (e.g. nonsmooth contact dynamics). You will build a robust framework to predict the multiscale behaviour of granular media with rigorous scale-bridging techniques based on stochastic homogenisation and benchmarked against available laboratory tests. The research position hinges on the interdisciplinary integration of AI with theoretical / computational mechanics and physics to deepen the scientific understanding and modelling of complex materials. The outcomes will contribute to advancements with impact in the scientific community and the industry. Avantages Subsidized meals Partial reimbursement of public transport costs Leave : 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.) Possibility of teleworking (90 days / year) and flexible organization of working hours Professional equipment available (videoconferencing, loan of computer equipment, etc.) Social, cultural and sports events and activities Access to vocational training Complementary health insuranceunder conditions Rémunération 2788€ gross salary / month #J-18808-Ljbffr