Understanding “If-Then” Relationships
Many global challenges, from climate change to healthcare and pandemic preparedness, involve systems where small changes can have far-reaching effects. Understanding how interventions influence outcomes in such complex dynamics requires reliable “if-then” reasoning. Traditional mathematical dynamical models often oversimplify these systems, while purely data-driven machine learning models, though powerful, can be difficult to interpret and may not generalize well to new situations. The DYNAMICAUS project, led by Niki Kilbertus, addresses this gap by combining machine learning methods with rigorous mechanistic modeling and methods from causal inference.
Hybrid Models and Uncertainty
The project focuses on hybrid dynamical models – mathematical frameworks that capture both physical knowledge and data-driven insights. By developing methods to quantify uncertainty and actively gather data where it matters most, DYNAMICAUS aims to reliably predict how different interventions will affect outcomes of interest in the future. This allows researchers to evaluate the potential impact of intervention policies in a more reliable and transparent way.
“By combining machine learning with causal inference in hybrid dynamical systems, DYNAMICAUS aims to deliver reliable insights that help address complex societal challenges in a responsible and impactful way,” so Kilbertus.
Applications in Climate, Health, and Epidemics
Niki Kilbertus and his team will apply their methods in areas of high societal relevance. In climate research, they aim to improve predictions of environmental interventions. In healthcare, the methods are designed to support treatment planning through better anticipation of patient outcomes. In epidemic simulations, the project strives to deepen understanding of intervention effects, helping to inform policy and preparedness strategies.
Ethics and Societal Impact
From the beginning, an ethicist will be integrated into the research process to address societal implications and guide responsible application of the methods. This ensures that technical developments are aligned with positive social impact and evidence-based decision-making.
