Céline Beji, PhD
I am a postdoctoral researcher at Paris Cité University, where I am collaborating with Prof. Raphaël Porcher in the Personalized Medicine Team of METHODS, CRESS UMR 1153, Inserm. My area of expertise is in statistical machine learning, focusing on Causal Inference in the counterfactual framework (Rubin causal model) and its applications to healthcare. My research focuses on individual and average treatment effects, risk/benefit classification, compliance and use of observational data. This research is a continuation of my thesis work, conducted at LAMSADE, Paris Dauphine-PSL University, under the supervision of Prof. Jamal Atif and Dr. Florian Yger as part of the MILES machine learning team.
I am also involved in Deeptech innovation and entrepreneurship. I am currently studying for a specialized master’s degree EDI at Mines-PSL, which is located in the dynamic ecosystem of Campus PariSanté. I am particularly interested in the role of research labs and startups in the development of healthcare innovation, as well as the ethical and regulatory aspects.
Causal Inference, Counterfactual, Treatment Effect
Deeptech Entrepreneurship and Innovation
Beji, C. (2021). Causal Populations Identification througth Hidden Distributions Estimation. Thèse de doctorat. [thesis]
Beji, C. & Yger, F.n & Atif, J. (2021). Non parametric estimation of causal populations in a counterfactual scenario. Causal-AI 2021. [paper]
Beji, C., Bon, M., Yger, F., & Atif, J. (2020). Estimating individual treatment effects through causal populations identification. ESANN 2020. [paper]