Cells and economies are examples of complex systems. In complex systems, individual components interact with each other giving rise to complex networks of interactions that are neither totally regular nor totally random. Partly because of the interactions themselves and partly because of the interaction's topology, complex systems cannot be properly understood by analyzing their constituent parts in isolation. This feature of complex systems poses important challenges from both a fundamental perspective and an engineering perspective. Roger's research is devoted to the study of complex systems, using probabilistic approaches taken from statistical physics.

During the last years, Roger's work has turned to exploring the intersection between these probabilistic approaches, Bayesian inference and machine learning. In this area, he has contributed novel methods for interpretable machine learning, and uncovered some fundamental limitations in model-learning.