Flood Risk Assessment using Remote Sensing Dataset and Machine Learning Algorithms - (F/H) - Internship

Location 75-PARIS, France
Experience level Internships
Job details sector Risk Management
Apply before Date not available

The AXA Group


The AXA Group, a worldwide leader in Financial Protection, supports and advises its customers, individuals and businesses at every stage of their lives by meeting their needs for products and insurance, pension, savings and wealth transmission.


The headquarters of AXA Group, based in Paris 8th, brings together the Group's corporate activities. It coordinates the various entities with the Group's strategy, and is responsible for managing international projects. The peculiarity of the seat lies particularly in the presence of a strong international culture (37 nationalities).

Aware of the challenges, AXA is a responsible and innovative Group that is constantly redefining standards to better serve its customers and employees.


The Departments


The internship will be based at the AXA Group Risk Management (GRM) department and developed in  partnership with the AXA Data Innovation Laboratory (DIL).


Monitoring natural catastrophe (CAT) risks is a key activity within the AXA GRM department. AXA GRM brings together multidisciplinary high-level teams composed of actuaries, engineers, PhDs and financial analysts, based in Paris (65 people), Zurich (15 people) and Madrid (20 people).

The Natural Hazards Team of AXA GRM is looking for an intern for 6 months. The intern will join a dynamic 12-members team dedicated to the modeling and monitoring of the CAT risks borne by the company.

The intern will conjointly work with the AXA Data Innovation Lab (DIL) R&D team for the design of the machine learning algorithms. The R&D team of the AXA DIL is dedicated to research activities on advanced machine learning techniques for AXA: for instance, deep learning, fair and transparent machine learning and machine learning on unstructured data. The team is composed of research data scientists (engineers and PhDs). It has strong relationships with academic labs in AI.


Role and Responsibilities :


Over the recent years, we have observed that flood risk and the associated damages are rising drastically. This increase mainly comes from two reasons. First, as climate is getting warmer, the atmospheric temperatures increase which amplifies atmosphere’s water holding capacity and the potential for extreme precipitation to develop. Second, the rapid urbanization and the wealth growth are enhancing flood risk by generating the development of new and more pronounced flood-prone areas.    


Examples of recent strong flood events are more and more numerous. Historic Hurricane Harvey caused in 2017 in Texas a 1000-year (0.1% chance of happening in any given year) flood event unprecedented in intensity. South Asian monsoon season in 2017 is now considered as one of the most intense and active season in decades in term of precipitation intensity and frequency bringing massive floods and associated landslides.   


New high-precision satellite datasets easily accessible nowadays provide directions to help to monitor such extreme climate events. Surface water can be observed using passive microwave remote sensing on-board instruments during and after an event. Such large-volume datasets contain raw measurement data that requires specific image processing treatments to be interpretable and useful. European Space Agency (ESA) Copernicus Sentinel-1 satellite data is a prime example of promising data for near-real time flood monitoring.


The intern will be first responsible for developing a typical classification model to identify surface water extent of extremes flood events. Secondly s/he will test and evaluate performance of several machine learning algorithms to derive flood water elevation depths. In addition, study of change detection using both pre- and post-disaster remote sensing images will be performed to build damage assessment.

This last point will be very useful for the calibration and validation of vulnerability modules -dedicated to the estimation of destruction rates for risk exposed assets- used in natural catastrophes models. 

Profile Requirements :
Master’s degree in computer science, mathematics or equivalent (engineering schools…) with interest for academic research. Candidates should have strong analytical mindset, strong programming skills (Python, R), as well as good communication skills to be able to work as part of a team.  
The candidate should have a strong background in machine learning. Experience in image processing, deep learning (DL) techniques for images and the ability to use DL frameworks are a plus.

Would you like to wake up every day driven and inspired by our noble mission and to work together as one global team to empower people to live a better life? Here at AXA we strive to lead the transformation of our industry. We are looking for talented individuals who come from varied backgrounds, think differently and want to be part of this exciting transformation by challenging the status quo so we can push AXA - a leading global brand and one of the most innovative companies in our industry - onto even greater things.

In a fast-evolving world and with a presence in 64 countries, our 165,000 employees and exclusive distributors anticipate change to offer services and solutions tailored to the current and future needs of our 107 million customers.


The headquarters of the AXA Group, based in Paris 8th, brings together the Group's corporate activities. It coordinates the various entities with the Group's strategy, and is responsible for managing international projects. The headquarters has approximately 800 employees and is distinguished by its strong international culture (39 nationalities).