On Mon, Aug 7, 2023, 9:46 AM James Bowery <[email protected]> wrote:

>
>
> On Sun, Aug 6, 2023 at 9:13 PM Matt Mahoney <[email protected]>
> wrote:
>
>> ...
>> A few years ago I researched homicide rates and gun ownership rates by
>> country and was surprised to find a weak but negative correlation. But that
>> doesn't tell us why. Does arming everyone deter crime, or does crime result
>> in stricter gun laws? The data doesn't say. You can use either data point
>> to predict the other and get the same compression.
>>
>
> For this kind of extremely limited causal analysis between just two
> variables, you might make progress with BMLiNGAM
> <https://taku-y.github.io/bmlingam/>, which is a package I used to create
> such paired causal inferences as a first step toward doing trivial path
> analysis on a bunch of data.  But, really, the proper approach is to go
> multivariate out of the gate.
>

I see that BMLiNGAM is based on the LINGAM model of causality, so I found
the paper on LINGAM by Shimizu. It extends Pearl's covariance matrix model
of causality to non Gaussian data. But it assumes (like Pearl) that you
still know which variables are dependent and which are independent.

But a table of numbers like LaboratoryOfTheCounties doesn't tell you this.
We can assume that causality is directional from past to future, so using
an example from the data, increasing 1990 population causes 2000 population
to increase as well. But knowing this doesn't help compression. I can just
as easily predict 1990 population from 2000 population as the other way
around.

As a more general example, suppose I have the following data over 3
variables:

A B C
0 0 0
0 1 0
1 0 1
1 1 1

I can see there is a correlation between A and C but not B. I can compress
just as well by eliminating column A or C, since they are identical. This
does not tell us whether A causes C, or C causes A, or both are caused by
some other variable.

What would be an example of determining causality with generic labels?

------------------------------------------
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/T772759d8ceb4b92c-Me179c26239d34b6412482b91
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Reply via email to