France’s National Biodiversity Plan published on 4 July 2018 has the overall objective of “Reconquering biodiversity, a question of survival for our societies”. One of the plan’s actions is to develop research and knowledge about biodiversity, with goals including defining integrative indicators to quantify and map the state of biodiversity.
Knowledge about biodiversity and its evolution is both a stimulating field for applying AI research and a major scientific and societal challenge. Artificial intelligence can make significant contributions with new dedicated solutions involving complex systems that incorporate heterogeneous elements and information.
The IA-Biodiv challenge thus aims to support Artificial Intelligence research in the field of biodiversity, by mobilizing cutting-edge French research at the convergence of the French National Artificial Intelligence Research Strategy and the France’s National Biodiversity Plan. The challenge is launched by the ANR (Agence Nationale de la Recherche) in partenership with AFD (Agence Française de Développement), relying on a donation from Facebook AI Research, for a global effort of 4.8 million €.
The IA-Biodiv challenge is intended for scientific communities in AI and biodiversity in France and AFD's partner countries in Africa. International consortia will be financed by Expertise France (mandated by AFD) according to the participation modalities for international consortia.
This call is open to two types of consortia:
The challenge presents a roadmap with defined and common data sets. The challenge focuses on Mediterranean and Pacific coastal marine environments (including the coastline). Data from other oceans/marine environments (e.g. Indian Ocean) are not part of the data provided and evaluated in this challenge. However, they can be added to the IA-BiodiNet common database and can be used by teams for their research, provided that they are made available to other participants via IA-BiodivNet.
The study of coastal marine environments has many advantages. First, Opensource data are numerous, and they meet the needs of an evaluation of artificial intelligence systems. In addition, essential ocean variables (EOV) have strong scientific and societal impacts. The IA Biodiv challenge, through the Operational Committee (COpé), is committed to providing these data for research and evaluations to the teams. The expertise, consultation and evaluations will be based on these data only. The scientific approaches developed for the challenge must be as generic as possible and transposable to other fields of application.
The four years of the challenge will be entirely and solely devoted to data from coastal marine environments. There are no plans, for the moment, for another AI-Biodiversity challenge on other data.
The IA-BiodivNet platform is intended, in the long term, to host all research data in biodiversity.
All datasets needed for the challenge evaluation campaigns will be provided. Data collection should not be part of the project expenses.
The Copé will make available, on this platform, only data from coastal marine environments in the Pacific Ocean and the Mediterranean Sea. Any contribution to the enrichment of this platform is encouraged throughout the challenge (data, tools, annotations).
A description of the data will be available on the challenge webpage in early summer 2021. The data, as such, will be made available to the participating teams at the beginning of the challenge (early 2022). The data will be enriched, by the Copé and the research teams, throughout the duration of the challenge.
Initially, the data provided by the challenge organization will be made available to the participating teams only. It will be made available to all research communities at the end of the challenge.
AI teams that have not previously worked on coastal marine data are perfectly eligible and relevant, regardless of the nature of the data they are working on.
It is preferable that biodiversity research teams have experience with coastal marine environments, but this is not required.
Please refer to the texts of the modalities for more details.
Please note that only French partners funded in consortia will have access to GENCI infrastructure.
Previous collaborations are not a prerequisite for submitting a proposal.
A website is at your disposal to search for potential partners or to submit an announcement.
You can send your questions to IA-Biodiv(at)anr.fr
The replay of the information webinars is available here.
We also encourage you to consult :
ANR's financial regulations
Text of the call
Participation modalities (French teams)
Terms of participation (French-African teams)
To be eligible, the lead applicant must meet the following conditions:
The lead partner and/or lead scientific coordinator (legal person) must be based in an eligible country. The lead partner and its partners must provide the identification documents for the legal person/legal entity (registration certificate/registration number).
The place and the legal existence of the project leader define where the entity is based, in accordance with the registration certificate/number provided.
Organizations based in a non-eligible country can be part of the consortium but no finance will be granted to the organization and will not have access to the GENCI database.
Yes, the funding for French partners is paid directly by Expertise France, after validation of the financial report by the consortium leader. If there are several French partners in the consortium, the funding for the French partners is paid directly by Expertise France to each partner.
Yes, the costs of employees can be covered by the Challenge's budget, if they do not have civil servant status or if they are not already paid by public funds (e.g. employees of a foundation, fixed-term civil servants, company employees, etc.).