News for hackers. Stuff that matters.

A comment on New method identifies the root causes of statistical outliers.

Re:New method identifies the root causes of statistical outl... by uoaei on Tuesday July 19, 2022 @03:41PM (#32156903)
This "new" method seems to be an old method, indistinguishable from Pearl's do-calculus.

What is the innovation here, aside from a new software library? The quantification of each candidate root cause's influence on the outcome? I am surprised the authors found nothing similar throughout the entire corpus of academic research.

I must say, Amazon's "science" blog is the most unimpressive of the big tech companies. It churns out PR like the rest, but the others at least have some substance behind them.

[ Parent | on HN ]
Re:New method identifies the root causes of statistical outl... by tpoacher on Tuesday July 19, 2022 @04:21PM (#32157414)
It's ironic that they say they find no good definition for the term, but then go on to cite the same people that use those exact terms to describe the exact same thing: https://www.semanticscholar.org/paper/Causality%2C-Responsib...
[ Parent | on HN ]
Re:New method identifies the root causes of statistical outl... by svnt on Tuesday July 19, 2022 @04:52PM (#32157800)
It would be more ironic if they made a mistake like this in a way that wasn’t quite so ego-reinforcing.
[ Parent | on HN ]
Re:New method identifies the root causes of statistical outl... by just_boost_it on Tuesday July 19, 2022 @07:56PM (#32159539)
Every one that I see is a pretty weak summary of a decades old idea, but presented like its new research.
[ Parent | on HN ]