Causal AI – Paul Hünermund, Ph.D.

A frequent point of criticism against Directed Acyclic Graphs is that writing them down for a real-world problem can be a difficult task. There are numerous possible variables to consider and it’s not clear how we can determine all the causal relationships between them. We recently had a Twitter discussion where exactly this argument popped up again.

I don't think there is such a thing as the "ground truth" in this context. In any context there are an enormous number of variables that could conceivably be included in a DAG, and which DAG you write will depend on whether you start with a tDAG or not. You said you disagreed.

— Jason Abaluck (@Jabaluck) September 4, 2019

Continue reading Mapping Unchartered Territory