Causal estimation of treatment effects in a federated environment.

Ponente:  Alejandro Almodóvar.
Organismo:  GAPS – SSR de la E.T.S.I. Telecomunicación, Universidad Politécnica de Madrid.

Fecha: 30 de Junio, 2023.
Hora: 10.00 horas.
Lugar: Sala B-223 de la Escuela Técnica Superior de Ingenieros de Telecomunicación.  [Cómo llegar]


Estimation of treatment effects from observational data is a challenging task for conventional machine learning methods, mainly because usually the distribution of trated and not treated patients is different. To deal with this problem, some machine learning techniques based in causal inference has been studied. The current specific task to be solved is to improve the treatment effect estimation in a federated setting, where different hospitals have data of their patients that cannot be shared between them.