[FRIAM] Judea Pearl: Book of Why

[Marcus Daniels]

I have a hard time with this as a way to extend data.   If it is high-dimensional it will be under-sampled.  Seems better to me to  measure or simulate more so that the joint distribution can be realistic.  And if you can do that there is no reason to infer the joint distribution because you *have* it.

On Apr 19, 2020, at 8:18 AM, Frank Wimberly <wimberly3 at gmail.com> wrote:


Going back and forth:  If you infer the causal graph from observational data you can use that graph to simulate data with the same joint distribution as the original data.


Frank

On Sun, Apr 19, 2020 at 9:11 AM uǝlƃ ☣ <gepropella at gmail.com<mailto:gepropella at gmail.com>> wrote:
The *ensemble* point is the primary reason I regret not being able to parse your response to my Necker cube summarization of EricS' TLDR. It goes back to the original question of how/whether distributional conceptions better catch the unknown unknowns left dangling in the ambience. Pearl's attempts to burst "causality" into graphs, away from chains (though helping to identify chains when they do exist) is along the same line.

To boot, it evokes both Gödel's interpretation of von Neumann's interpretation of Gödel's work (that it takes an infinite expression to describe a thing) and Rosen's definition of complexity (basically anything that requires an infinite number of models to describe).

And, although I can't get my hands on the Rota paper EricS posted, I'm leery of relying on any phenomenology. Heidegger I trust a bit. Husserl not so much. Regardless, I don't think it's *necessary* to go that deep to grok the main point, which is that the transformation should be invertible. We should be able to flip back and forth from goo to thing such that the flipping doesn't change it. The goo we get after flipping from the things should be the same goo we had to start with.

On 4/19/20 6:25 AM, Steven A Smith wrote:
> My work of late (other than SimTable) has been in the realm of trying to analyze ensembles of predictive simulations.   This is a logical next step (forward and backward propogating data and constraints as they are recorded/discovered/postulated) across space (populations) and time.


--
☣ uǝlƃ

.-. .- -. -.. --- -- -..-. -.. --- - ... -..-. .- -. -.. -..-. -.. .- ... .... . ...
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam<http://bit.ly/virtualfriam>
unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
archives: http://friam.471366.n2.nabble.com/
FRIAM-COMIC<http://friam.471366.n2.nabble.com/FRIAM-COMIC> http://friam-comic.blogspot.com/


--
Frank Wimberly
140 Calle Ojo Feliz
Santa Fe, NM 87505
505 670-9918
.-. .- -. -.. --- -- -..-. -.. --- - ... -..-. .- -. -.. -..-. -.. .- ... .... . ...
FRIAM Applied Complexity Group listserv
Zoom Fridays 9:30a-12p Mtn GMT-6  bit.ly/virtualfriam
unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
archives: http://friam.471366.n2.nabble.com/
FRIAM-COMIC http://friam-comic.blogspot.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20200419/47225872/attachment.html>