Hey gts,

I think this topic is more appropriate for

[email protected]

which you can sign up for at the same place as you signed up for this
Singularity email list.

The reason is that foundations of probability is a highly technical
issue of relevance to AGI engineering; whereas this email list is
intended more for general discussion of Singularity-related issues...

Anyway, to respond to your point: Yep, I agree that exchangeability is
different from, but closely related to Chaitin randomness, in the
sense that for finite series it seems to be the case that

* Chaitin randomness "almost always" implies exchangeability
* Exchangeability "almost always" implies Chaitin randomness

-- Ben

On 1/19/07, gts <[EMAIL PROTECTED]> wrote:
Ben wrote on ExI:

> However, exchangeability likely does NOT imply Chaitin randomness....

I think you're right about that, though I confess my understanding of
Chaitin randomness is still sketchy.

My understanding is that exchangeability in the de Finneti sense implies
only the sort of randomness defined by the more conventional axiom of
randomness, which I sent to you and a few others in private email this
morning:

"Axiom of Randomness: the limiting relative frequency of each attribute in
a collective C is the same in any infinite subsequence of C which is
determined by a place selection."

(http://plato.stanford.edu/entries/probability-interpret/)

This is only the ordinary frequentist view of randomness.

De Finneti viewed exchangeability as a reduction of the objectivist notion
of independence, i.e., on his view we can and should reduce the allegedly
metaphysical notions of 'objective probability' and 'independence' to his
equivalent notions of 'subjective probability' and 'exchangeability'.

Nothing there as far as I know about the sort of maximum-entropy
randomness that I think incompressible Chaitin-random sequences are
thought to be.

-gts

-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?list_id=11983


-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?list_id=11983

Reply via email to