The Artificial intelligence for science, science for artificial intelligence (AISSAI) center

As part of its strategic plan on Artificial Intelligence (AI), the French National Center for Scientific Research (CNRS) has launched its AI for Science and Science for AI (AISSAI) Center. The main objective of this center is to structure and organize transverse actions including all CNRS institutes that interact with AI.

logo d'AISSAI

 

International cooperation

There are several schemes for structuring international collaborations, including IRL, IRP and IRN, with dedicated funding. Other initiatives, such as the IEA, also support scientific exchanges.

Depending on the maturity and history of the proposed cooperation, various types of mechanism can be considered:

  • International Research Laboratories (IRLs) structure the significant, long-term presence of scientists from a limited number of French and foreign research institutions (one foreign partner country only) in a single, identified location. They are fully-fledged research units of the CNRS.
  • International Research Projects (IRP) are collaborative research projects set up between one or more CNRS laboratories and laboratories in one or two foreign countries. They are designed to consolidate established collaborations through short or medium-term scientific exchanges. Funding ranges from €10,000 to €15,000 per year for 5 years.
  • International Research Networks (IRN) bring together several French and foreign partners to focus on a specific theme. Funding ranges from €10,000 to €15,000 per year for 5 years.

There is no call for the creation of IRLs. This is discussed directly at CNRS Informatics in conjunction with the International Department.

For IRNs and IRPs, a call for interest (AMI) is launched annually. You will find here the documents corresponding to this AMI :

However, there are other types of action that promote international cooperation via regular calls:

  • Numerous International Emerging Actions (IEA) also encourage partnerships and the exploration of new fields of research.
  • There are also numerous international programs, such as bilateral ones, relayed by the DEI.

Fields of Research

Artificial intelligence (AI) is a field of research in its own right, but it can also accelerate scientific discovery in many fields, provided that it fosters closer ties between different scientific communities. Because of its national and multidisciplinary scope, as well as its international standing, the French National Center for Scientific Research (CNRS) must play a driving role in promoting these interactions between different fields of research. As they are not only related to the assimilation of AI methods (in particular, machine learning), they are also necessary for the removal of barriers specific to this field related to strategic subjects such as explainability, the incorporation of expert knowledge, and frugality.

AI is certainly transforming our societies, but AI, as a knowledge discovery paradigm, is also impacting the way research is produced across disciplines. AI methods, in particular Machine Learning, represent today an important challenge for modeling increasingly complex and multiscale systems, as well as exploiting the massive increase in flows, volumes and the diversity of multi-source data from large instruments, observation systems, experimental platforms, etc. They are leading to new research practices whose level of maturity and organization varies greatly from one discipline to another, as well as new needs in terms of expertise and international collaborations.  Within the CNRS, AI methods are now being deployed very rapidly in the context of major research infrastructures and national and international experimental platforms that the CNRS supports or to which it contributes (for instance, in biology, high-energy physics, climate modeling, etc.). The very aim of the AISSAI Center is to foster dialogue between disciplines, to address new scientific questions and to establish radically new modes of collaboration between scientific fields.