gts,

About the Principle of Indifference and probability theory...

The question is what should an AGI system does when the data available
to it appears to support multiple contradictory conclusions.

It has to decide, somehow.

The PI is one way to decide...

Another approach however is the Occam prior (to prefer simpler
explanations, according to some fixed measure of simplicity, provided
e.g. by assumption of some reference universal turing machine).

I think that in general the Occam prior is preferable to the PI as an
assumption.

I note that the Occam prior connects more closely to neuroscience than
the PI, in that there are plausible arguments the brain uses an
"energy minimization" heuristic in some cases.  Read Montague makes an
argument in this direction in:

http://www.amazon.com/Why-Choose-This-Book-Decisions/dp/0525949828

However, when multiple choices seem to have roughly equivalent
complexity, then the Occam prior basically degenerates to the PI.

There are mathematical relationships between algorithmic information
and Shannon entropy, which indicate that the Occam prior and the
maxent prior are equivalent in some cases; e.g.

http://prola.aps.org/abstract/PRA/v40/i8/p4731_1

which shows

"Algorithmic randomness is typically very difficult to calculate
exactly but relatively easy to estimate. In large systems,
probabilistic ensemble definitions of entropy (e.g., coarse-grained
entropy of Gibbs and Boltzmann's entropy H=lnW, as well as Shannon's
information-theoretic entropy) provide accurate estimates of the
algorithmic entropy of an individual system or its average value for
an ensemble."

Zurek argues, further, that

"
Physical entropy, I suggest, is a sum of (i) the missing information
measured by Shannon's formula and (ii) of the algorithmic information
content—algorithmic randomness—present in the available data about the
system.
"

A clear, freely available exposition of Zurek's views is found here:

http://arxiv.org/pdf/quant-ph/9807007

This suggests that the physically correct prior might be a combination
of maxent and Occam priors.

The PI then emerges as a kind of degenerate version of this more
sophisticated approach.

And, just as with the PI, these more sophisticated approaches must be
applied correctly and intelligently to be useful.

-- Ben

On 2/10/07, gts <[EMAIL PROTECTED]> wrote:
On Sat, 10 Feb 2007 15:27:23 -0500, Jef Allbright <[EMAIL PROTECTED]>
wrote:

> On the contrary, a subjectivist understands that even to pose a
> question, one must have some prior.

That observation does not speak to the question at hand concerning the
principle of indifference.

The principle of indifference is seen as a 'logical principle' only under
objective bayesianism. Under subjective bayesianism it is at most a
heuristic device.

Subjectivists know better than to believe they are bound by some
'universal principle of logic' (to use your term) to invoke the principle
of indifference under conditions of total ignorance about the true state
of nature, which is of course the only condition under which it can be
invoked.

You were wrong to suggest earlier that the principle of indifference can
be derived from De Finetti coherence. The axioms of probability can be
derived from coherence but the principle of indifference is certainly not
one of them.

> You're also confusing "zero" with "nothing."

Nope.

-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=303


-----
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=303

Reply via email to