The process of automatically synthesizing models from the data for the purpose of prediction and control is the methodological core of contemporary science and engineering. While different fields have their own ways to collect and process data, the inference problems that need to be solved often share common underlying structures. Abstracting out such structures and formulating flexible algorithmic frameworks able to process disparate types of data is a fundamental step towards boosting interdisciplinary exchange.

My research aims to develop inference and control methodologies that are theoretically sound and exhibit optimality, robustness, and scalability properties. My approach is to formulate these questions in the form of well-posed optimization problems whose solution can be computed efficiently. More generally, I am interested in methodological and applied developments in the fields of statistical inference, machine learning, non-linear control, identification of dynamical systems, and large scale optimization.

For a short biography and a list of publications, please have a look at my personal web page.