arXiv:1305.5506cs.AI
This is a purely pedagogical paper with no new results. The goal of the paper is to give a fairly self-contained introduction to Judea Pearl's do-calculus, including proofs of his 3 rules.
Linked by 2 identifiable commenters in 3 comments across 2 threads, from Jan 2015 to Oct 2023.
This is an interesting illustration of correlation vs causality. By "causality", I mean links which remain even after you change the composition of your inputs. See the do calculus, which is one of the better attempts to codify it: http://arxiv.org/pdf/1305.5506v1.pdf…
Connected through shared HN threads, not inferred similarity or a recommendation.
This modelling ambiguity is resolved by do calculus, which makes a clear distinction between intervention and observation: https://arxiv.org/pdf/1305.5506.pdf
NatsuView on HNParent context
There is something missing: regular stats don't differentiate between doing things and observing things and these two are not at all the same. If I have a digital thermometer and I observe it to show a high temperature, then I will note an association between that and feeling…
NatsuView on HNParent context