Talk Title: In Search of Lost Time Ordering
Perhaps the least controversial claim one could make about causation is that causes typically precede their effects. For this reason, it is surprising that time does not play a role in the semantics of graphical causal models. This puzzling fact reflects a significant and unappreciated feature of causal models. Namely, many causal models are implicitly relativized to the spatiotemporal scale at which the target system is represented. The way that causal inference methods learn causal ordering without time ordering is by exploiting the independence relations that arise due to the dominant behaviors at distinct scales. This scale-relativity of causation has wide-ranging implications for understanding the worldly infrastructure underlying causal attributions, for mapping the relationship between causal and dynamical representations, and for the design of policy interventions.
Naftali Weinberger
Naftali Weinberger is a scientific researcher at the Munich Center for Mathematical Philosophy. His work concerns the use of causal methodology to address foundational questions arising in the philosophy of science as well as questions arising in particular sciences, including: biology, psychometrics, neuroscience, and cognitive science. He currently has two primary research projects – one on causation in complex dynamical systems and another on the use of causal methods for the analysis of racial discrimination. He is currently trying to convince causal researchers that causal representations are implicitly relative to a particular time-scale and that it is therefore crucial to pay attention to temporal dynamics when designing and evaluating interventions.