Junior Research Scientist in machine learning for spatial statistics 

Back to campaign's jobs listing

INRAE presentation

The French National Research Institute for Agriculture, Food, and the Environment (INRAE) is a public research establishment under the dual authority of the Ministry of Agriculture and the Ministry of Research.

INRAE is recruiting researchers by open competition and offering permanent position.

It is a major player in research and innovation created on 1^st of January 2020. INRAE is a research institute resulting from the merger of INRA and IRSTEA. It is a community of 12,000 people with more than 200 research units and 42 experimental units located throughout France.

The institute is among the world leaders in agricultural and food sciences, in plant and animal sciences, and is 11^th in the world in ecology and environment. INRAE’s main goal is to be a key player in the transitions necessary to address major global challenges. In the face of the increase in population, climate change, scarcity of resources and decline in biodiversity, the institute develops solutions for multiperformance agriculture, high quality food and sustainable management of resources and ecosystems.

Work environment, missions and activities

The Biostatistics and Spatial Processes (BioSP) Research Unit at INRAE (Avignon) develops research in the fields of statistics, dynamic systems, ecology-epidemiology with a focus on spatial and spatio-temporal problems. Major application domains include ecology, epidemiology, agriculture, climate and the environment. Spatial and spatio-temporal statistics (Gaussian processes, extreme processes and point processes) are historically at the core of research at BioSP, and today they constitute one of its main disciplinary axes. The assessment of epidemiological, environmental or climate-related risks can be improved by capitalizing on the exponential growth in the number and volume of databases, an evolution which impacts both spatial analysis methods (due to dimensionality) and Machine Learning methods (due to spatial and spatio-temporal dependencies). Hybridizing these two approaches and their relative strengths is thus necessary and represents a major scientific challenge. Ultimately, the ambition is to modernize the toolkit of spatial statistics and to position BioSP as an actor at the forefront of theoretical and methodological contributions in this field of research.

As successful candidate, you will become part of this research axis at BioSP and of the Division of Mathematics and digital technologies at INRAE. You will develop research in the field of machine learning for data with spatial and/or spatio-temporal dependencies arising in environmental, climatological and ecological studies. Since this research field is vast with numerous research avenues and opportunities, you will have the autonomy to define and develop your research priorities within this perimeter. Through your theoretical experience and contributions, you will renew and extend approaches in spatial statistics and strengthen the research team at BioSP by fostering the use of Machine Learning and Deep Learning techniques. You will collaborate with BioSP members in the fields of spatial statistics, applied Machine Learning, extreme events and epidemiology. You will also take part in the projects developed within the scope of the Geolearning Chair jointly coordinated with the Geostatistics team at Ecole de Mines (Paris).

In line with INRAE's policy on open science, you will share your research work with the scientific community through publications and make R/Python packages available for wide dissemination. Your work will build on the rich network of collaborations already established by BioSP and expand it at different levels: local, national and international. You will be involved in training activities (Masters, research schools) and supervision of Master students, PhD candidates and postdocs.

Our research team tells you more about your future job

Training and skills

You hold a PhD degree or equivalent.
Skills and experience in Machine Learning methods (supervised, semi-supervised, unsupervised learning) are highly recommended. You have acquired knowledge and experience in modern deep learning paradigms such as domain adaptation, transfer learning, weakly supervised learning or knowledge distillation, GANs, deep generative models, including the development of these in a data analysis and processing framework.
You have shown your ability to prove novel mathematical results, for example on theoretical guarantees for these methods.
You know how to communicate your results and make them operational through open computer codes.
Moreover, already proven experience in the analysis of spatial data, especially in the fields of climate, environment and/or ecology, would be appreciated.
You enjoy teamwork and you have good interpersonal skills. You show initiative and autonomy.
Fluency in English and knowledge of French are desired, as is long-term international experience: successful candidates who have not yet gained such experience will be strongly encouraged to make a stay abroad at the end of the internship year, prepared jointly with the host team.

INRAE's life quality

By joining our teams, you benefit from:

- 30 days of annual leave + 15 days "Reduction of Working Time" (for a full time);
- parenting support: CESU childcare, leisure services;
- skills development systems: training, career advise;
- social support: advice and listening, social assistance and loans;
- holiday and leisure services: holiday vouchers, accommodation at preferential rates;
- sports and cultural activities;
- collective catering.

For international scientists: please visit your guide to facilitate your arrival and stay at INRAE.