Postdoctoral Fellow in Statistical Machine Learning

The project's goal is to research and develop novel statistical machine learning and inference methods for predictive and prescriptive tasks that overcome the challenges posed by applied settings (e.g., non-stationary and missing not-at-random phenomena), combining probabilistic models, deep learning, stochastic processes and approximate inference.
The postdoctoral candidate will have the opportunity to pursue their own ideas, and to lead the design and implementation of novel machine learning solutions within the following research challenges:
1. How to disentangle data missingness patterns from time-varying signals of interest via statistical machine learning models.
2. How to devise data-driven, automated sequential decision making algorithms in dynamic and multi-scale prescriptive environments.
The candidate will investigate both the theoretical and practical aspects of statistical machine learning, with potential assessment of how the devised methods perform in healthcare applications and online digital services.
The candidate will work under the supervision of Ikerbasque Research Fellow Iñigo Urteaga, and will benefit from close collaboration with members of the Machine Learning group at BCAM and external collaborators at Columbia University.