Abstract:
Understanding the relationships of causes and their effects is crucial to any science, including medicine where precise treatment recommendations can make crucial differences for patient outcomes. Causal Inference helps quantify these relationships with the help of statistical modelling. In this talk, I highlight the value of bounds on causal effects as a stepping stone towards more robust causal analysis. More robust analysis helps justify results in classical biomedicine studies using methods such as Mendelian randomisation, IV and more. Through that, bounds will play a crucial role in making those results more accessible, impactful and ultimately, more successful.

Bio:
Jakob Zeitler researched causal inference under the supervision of Ricardo Silva at University College London during his PhD in Foundational Artificial Intelligence, funded by Google Deepmind. During his PhD he interned at Spotify Research, yielding one paper and two patents. At Oxford, he serves as a Pioneer Fellow in the SMARTbiomed centre, joining Robin Evans group at the Department of Statistics to advance methods of causal inference, machine learning and combined.