���Dear SIPTAnts,

I���d like to draw your attention on a paper of ours recently accepted at PGM: http://people.idsia.ch/~zaffalon/papers/2020pgm-scmcns.pdf.

The paper shows that a structural causal model can be understood, and converted, into a (strong-independence-based) credal network. Whence it can eventually be solved by common algorithms for credal nets - in fact we are also providing a library for the conversion/solution: https://ipg.idsia.ch/preprints/cabanas2020a.pdf.

To me this looks like a nice opportunity for having imprecise probabilities more broadly taken into proper consideration, especially in communities that have had a lukewarm approach to it so far: in particular because nowadays causal inference is getting a strong momentum - other than being a beautiful and an important topic on its own (my viewpoint).

Hoping you will find the connection interesting and wishing all the best to you and your families in these weird times,
Marco