CNRS-Momentum : appel à projets 2017

Institutionnel

Le CNRS lance un appel à projets visant à permettre à de talentueux jeunes scientifiques indépendants d'imaginer et de mener à bien un projet innovant au sein d'un laboratoire du CNRS. En développant leurs propres projets, les lauréats du CNRS-Momentum contribueront à étoffer la recherche des unités qui les accueilleront.

Cet appel est ouvert, sans condition de nationalité, aux chercheurs titulaires du CNRS et aux non permanents, ayant soutenu leur doctorat (ou équivalent) depuis moins de 8 ans (doctorat après le 31/10/2009)1 .

En 2017, le programme CNRS-Momentum soutiendra des projets dans les domaines émergents et transdisciplinaires suivants (voir ci-dessous):

  1. Etude des cycles du carbone: des bio-pompes à l'économie circulaire
  2. Traitement de l'information par le cerveau: déchiffrage du code neuronal
  3. Stabilité et plasticité des Systèmes Complexes
  4. Inspiration et mimétisme
  5. Surfaces et interfaces
  6. Comportement humain, sur le plan collectif et social
  7. Sciences participatives: modèles, méthodes et outils
  8. Sécurité des données et transparence des algorithmes
  9. Nouvelles frontières de l'apprentissage automatique dans le domaine de l'intelligence artificielle
  10. Réseaux intelligents
  11. Modélisation du vivant
  12. Matériaux multifonctionnels : de l'échelle nanométrique à la description multi-échelle
  13. Calculs et simulations quantiques

Sélection

Les critères de sélection seront basés sur la qualité du candidat, ainsi que sur l'originalité et la pertinence de son projet par rapport aux thèmes choisis. Le choix des propositions reviendra au collège de direction du CNRS, présidé par Alain Fuchs, Président du CNRS.

La sélection s'effectuera en deux étapes: une présélection en octobre 2017, suivie d'entretiens des candidats présélectionnés en novembre 2017. La liste finale des lauréats sera établie courant novembre pour un début de financement en janvier 2018.

Financement

La bourse CNRS-Momentum sera allouée pour une période de 3 ans.

  • Financements équipements et fonctionnements à hauteur de 60.000 € maximum par an
  • Deux ans de salaire pour un post-doctorant ou un an de salaire pour un technicien
  • Trois ans de salaire pour les lauréats non titulaires

Le programme CNRS-Momentum n'est pas accessible aux jeunes chercheurs titulaires d'une bourse de recherche similaire (ATIP-Avenir, ANR JCJC ou bourse ERC - Starting/Consolidator).

Thèmes de recherche

  • 1La période effective écoulée depuis l'obtention du doctorat sera réduite d'un an en cas de congé maternité (ou davantage, sur présentation de justificatifs, selon la durée du congé pris à chaque naissance postérieure à l'obtention du doctorat), congé paternité, maladie de longue durée ou service national, sur présentation de justificatifs.

1. Investigating carbon cycles: from biopumps to circular economy

In order to cope with climate change, solutions enabling carbon capture are the next challenges to address. Living organisms are key drivers of the biogeochemical carbon cycle for the conversion into organic compounds. Investigating the adaptation of the complex processes involved in these biological pumps of carbon to a broad array of environmental constraints (marine and terrestrial) represent a major challenge. Mineralisation or hydrogenation processes (chemical or physicochemical transformation of CO2) are the early stages towards a CO2-based circular economy. These transformations need to be approached in a varied way - theory or experiments - to enable a comparative evaluation of different strategies.

Keywords: CO2, photosynthesis, carbon sequestration, biogeochemical cycle, climate, circular economy, process, mineralization

2. Information processing in the brain: cracking the neural code

The human brain is an outstanding system of information processing. However, understanding how the brain actually works, how it can perform such accurate processing with such inaccurate components are great challenges, with impact both in Neurosciences and Computer Sciences. The objective is to make a paradigmatic shift and break the neural code in a pluridisciplinary team.

Keywords: Neural code, brain information processing, brain modeling, brain signal processing and control, neuromorphic computing

3. Stability and plasticity of Complex Systems

Cell fate and collective cell behavior imply distinct processes, such as proliferation, specialization, trans-differentiation, movements, regeneration, ageing, etc. to essentially create a complex system stabilized and controlled by communication means. At a higher scale, communication and interaction also orchestrate the formation and the stability of resilient ecosystems. The grammar of chemical effectors underlying these processes, their control and environmental effects remain poorly understood but are mandatory to get new insight into the communication at the intra- and inter-cellular, intra- and inter-species levels in ecosystems and within populations.

Keywords:plasticity, robustness, chemical communication, resilience

4. Inspiration & mimicry

Natural systems and their unique properties are a powerful reservoir of abundant inventiveness that deserves to be described, analyzed and understood. They represent a source of inspiration for developing novel approaches, to tailor or shape new molecules, materials or composites with enhanced or unique properties, store or release energy, to settle new processes, to design and manufacture new autonomous devices or systems, bio-inspired robotic systems, information processing architectures or calculation paradigms. These approaches intrinsically associate biologists, chemists, physicists, mathematicians, computer scientists, or engineers in a pluri- or inter-disciplinary context.

Keywords: bio-inspiration, bio-mimicry, complex systems, modeling, analysis

5. Surfaces and Interfaces

Interfaces between solids, liquids or gases are the place of unique chemical, physical or biological processes often driven by the properties of involved surfaces. Surfaces can be stabilized, decorated, or functionalized to introduce tailored structural or dynamical properties. Interfaces can stabilize molecules, ensembles of molecules or nano-objects controlling exchanges or establishing compatibility or discontinuities between domains.

Keywords: Surfaces, Interfaces, Functionalization, Compatibility, Catalysis, Biointerfaces.

6. Collective and social human behavior

The analysis of the collective and social dimensions of human behavior is essential for understanding social dynamics and requires new formal analyzes through collaborations between social psychology, data sciences, cognitive sciences, economics, mathematics, anthropology, and sociology. The aim is to identify the relevant social interactions concerning issues such as (but not limited to) violence, discrimination, public health policies, labor and business developments, welfare, and education. The theoretical predictions should be tested empirically.

Keywords: behavior analysis, scientific computing and modelling, cognitive flocking models, social interactions, public policy evaluation.

7. Participatory Science: Models, Methods, and Tools

Participatory science is recognized as an original and fruitful approach to many problems such as large data sets acquisition, image and text annotation and interpretation, data aggregation, model construction and many other scientific goals. The research to be carried out is at the interface of social sciences and digital science, with a particular focus on players profiling, process modeling, subtasks allocation, coordination and synchronization, as well as crowd recruitment and rewards, legal and ethical issues. The research may benefit from multiple technologies such as web and human-machine interaction, serious games, process engineering and big data science.

Keywords: software engineering, serious games, community-based research, human-machine interaction, social rights, crowdsourcing, digital labor and teleworking.

8. Data Trust and Algorithms Transparency

Data trust and algorithms transparency have become crucial issues, addressed at the interface between information sciences and human and social sciences. Fundamental research must be carried out to define formal languages that describe algorithms and privacy rules in an intelligible way, and to provide auditing methods and monitoring tools that assist experts in evaluating unsatisfactory features and identifying liabilities with respect to ethical and legal rules.

Keywords: specification languages, behavioral models, privacy and security, algorithms auditing, data quality.

9. New frontiers of machine learning for Artificial Intelligence

Mathematics and computer science form the core of the fast growing area of large-scale machine learning. The focus is on the theoretical foundations, the treatment of massive data sets, including acceptability and privacy issues, and the application to real-world problems, including the derivation of the laws of nature. Specific areas of interest are: deep learning, representation learning (manifolds and metric learning, sparse coding and overcomplete representations, structured prediction,…), reinforcement learning, online learning and game theory, transfer learning, unsupervised learning, causal modelling, and stochastic and non-convex optimization.

Keywords: machine learning, large-scale problems, massive datasets, optimization, artificial intelligence.

10. Smartgrids

Smartgrids encompass research and technology concepts that allow societal challenges such as Climate change and Energy Transition to be addressed at the interface of new energy technologies, ICT and social and human sciences. The scientific challenges linked to the development of smartgrids concern several interdisciplinary aspects such as the integration of non-controllable and non-dispatchable variable generation (by massive integration of ENR) and loads (such as the development of the electric vehicle connected to the grid), the growing complexity of operation of large systems, coupled heterogeneous models of energy and ICT infrastructures with embedded intelligence, stochastic long term planning models taking into account responsibility partitioning, integration of energy storage, etc. Locks also exist with regard to observation and control of grids and modeling of very large systems with chaotic behaviors that are difficult to predict, or concerning the management of complexity in critical and coupled infrastructures. Finally, with regard to the uses, it is essential to collaborate in the design / planning and in the analysis of the advanced functions of network management, in order to obtain a balance between the technical functions related to the conduct of the grid and the economic aspects without degrading its security.

Keywords: Smartgrids, energy distribution, automatic control, optimization, simulation.

11. Modeling of living systems

Novel experimental methods in life sciences call for modelling in order to obtain quantitative insights that cannot be provided by experimental studies alone. A special attention will be deserved to candidates having already experienced interdisciplinary works with mathematics, scientific computing, biology, ecology or medicine, and developing rigorous analysis.

Keywords: cell assemblies organization, evolutionary biology, mechanisms of disease progression and therapy, living tissues, multi-scale analysis.

12. Multifunctional materials: from the nano-scale to a global multiple scale description

A major challenge to understand and design materials integrating several functionalities is to be able to link the materials properties from the nano to the macroscopic scale. Multiscale approaches can link atom-level mechanisms to the global physical, mechanical and chemical properties and their validation should integrate both experimental and theoretical state-of-the-art analysis techniques.

Keywords: multifunctional (nano-) materials, in-situ/in-operando characterization, multiscale analysis and modeling, asymptotic analysis, macroscopic models.

13. Quantum calculations and simulations

Issues related to quantum calculations raise the need for precise mathematical formulation of complex quantum systems. Going below standard asymptotic limits is necessary to achieve engineering and control of related processes. Systems with few degrees of freedom may be used to predict phenomena in more complex systems.

Keywords: open quantum systems, few-body systems, quantum algorithms, relativistic systems, derivation of effective models.