Post-Doctoral Research Visit F/M Policy learning under distributional shifts

The Inria centre at Université Côte d'Azur includes 42 research teams and 9 support services. The center’s staff (about 500 people) is made up of scientists of different nationalities, engineers, technicians and administrative staff. The teams are mainly located on the university campuses of Sophia Antipolis and Nice as well as Montpellier, in close collaboration with research and higher education laboratories and establishments (Université Côte d'Azur, CNRS, INRAE, INSERM ...), but also with the regional economic players.

With a presence in the fields of computational neuroscience and biology, data science and modeling, software engineering and certification, as well as collaborative robotics, the Inria Centre at Université Côte d'Azur  is a major player in terms of scientific excellence through its results and collaborations at both European and international levels.

Assignments :
With the help of Julie Josse, the recruited person will be taken to conduct innovative research on causal inference.

For a better knowledge of the proposed research subject :
A state of the art, bibliography and scientific references are available at the following URL, do not hesitate to log in the Premedical website

Collaboration :

The recruited person will be in connection with Aurelien Bellet, Research Director at Inria Montpellier. 

The recruited person will be part of the PEPR SMATCH (Statistical and AI based Methods for Advanced Clinical Trials CHallenges in Digital Health) in digital health and will beneficiate from this network. In particular, he/she will be part of workpackage 2: Enriching clinical trials with multi-sources and multidimensional auxiliary data. The aim is to develop the next generation of methods to combine high-throughput, high-dimensional individual and ancillary data in clinical trials and to evaluate their benefit.

Responsibilities :
The person recruited is responsible for developping methods of data integration in the context of causal inference  and will take initiatives for studying statistical properties of policy learning methods under distributional shifts. 

The inclusion/exclusion criteria of an RCT are closely tied to its ability to reach useful conclusions: across an overly heterogeneous group will kill statistical power on the average effect, but an overly homogeneous group risks misrepresenting the target population. Data fusion, drawing the link between the RCT and individuals outside the trial (Colnet et al. 2021), can address both these problems, in particular by estimating heterogeneous effects, rather than concluding on the average effects. The objective of this project is to use machine-learning methods to estimate heterogeneous effects from an RCT leveraging external data to increase statistical power. This project will provide concrete procedures and recommendations. A theoretical and numerical study will conclude on the finite-sample bias and variance of various machine-learning methods to estimate the CATE (conditional average treatment effect). The same methodology will be applied for time-to-event data. More precisely, we will go through the following steps

• Developping transportability estimators for policy learning and studying their statististical properties (asymptotic and finite sample) in particular with different subsets of variables
• Incorporate a temporal aspects and study dynamic treatment regimes
• Testing the suggested methods using numerical simulations and clinical data

In terms of concrete applications, PreMeDICaL has ongoing collaborations with
hospitals and other clinical partners. These collaborations will provide opportunities to apply the
approaches developed during the Postdoc to concrete use-cases. 

Technical skills and level required : PhD in Statistics, Machine Learning,  biostatistics or related fields. Strong statistical computing skill. 

Languages : English, French

Relational skills : Excellent writing and communication skills

• Subsidized meals
• Partial reimbursement of public transport costs
• Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
• Possibility of teleworking and flexible organization of working hours
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities
• Access to vocational training
• Contribution to mutual insurance (subject to conditions)

Gross Salary: 2788 € per month