More simply even that that, Pei, when it comes across a task and a choice of
options, if it sees no benefit 5% (arbitrary setting or 0%) does your system
choose randomly between between the choices?
Doesnt this make the system non-deterministic...
Otherwise agree with your description.
Mike Tintner [EMAIL PROTECTED] wrote: Pei,
I don't think there's any confusion here. Your system as you describe it IS
deterministic. Whether an observer might be confused by it is irrelevant.
Equally the fact that it is determined by a complex set of algorithms applying
to various tasks and
That would indeed be free, nondeterministic choice, which, as I understood, Pei
ruled out for his system.
The only qualifications are:
* choosing randomly is only one of an infinity of possible methods for such
choice
* the difference between options can be much greater than 5% - humans and,
I should have added -- the difference between options can be much greater
than 5% - humans and, offhand, I imagine, most AGI's, couldn't begin to
measure and compare options, with that degree of precision... for most
decisions (not, of course, all)
-
This list is sponsored by AGIRI:
Mike Tintner [EMAIL PROTECTED] wrote: That would indeed be free,
nondeterministic choice, which, as I understood, Pei ruled out for his system.
The only qualifications are:
* choosing randomly is only one of an infinity of possible methods for such
choice
rephrase this one?
*
On 5/8/07, James Ratcliff [EMAIL PROTECTED] wrote:
More simply even that that, Pei, when it comes across a task and a choice of
options, if it sees no benefit 5% (arbitrary setting or 0%) does your
system choose randomly between between the choices?
It depends on the type of the task.
If
Mike and James:
Please see my reply under a new subject, which also addressed the later posts.
Pei
On 5/8/07, Mike Tintner [EMAIL PROTECTED] wrote:
I should have added -- the difference between options can be much greater
than 5% - humans and, offhand, I imagine, most AGI's, couldn't
YKY (Yan King Yin) [EMAIL PROTECTED] wrote: On 5/7/07, James Ratcliff
[EMAIL PROTECTED] wrote:
One goal or project I was considering (for profit) is a research tool,
basically a KB that scans in teh newspapers and articles and extracts
pertinent information for others to query against and
James: is non-determinsm USEFUL? other than as I have stated, choosing
randomly when we dont know any better? Is there any other way to implement
non-determinsm, and is their any use for it?
Check this out - you are, I suggest, working on the assumption that
deterministic is reasonable, and
Ben, Pei,
How does your system handle choices such as this?
When given a fork can you return a % or number value back about which choice is
the best?
How finely graded does this get?
I believe simplisticly the blocks world example has to have a value function
when it calls something like
James,
For this level of details, you'll need to read my technical writings,
such as Confidence as Higher-Order Uncertainty
(http://nars.wang.googlepages.com/wang.confidence.pdf).
Pei
On 5/8/07, James Ratcliff [EMAIL PROTECTED] wrote:
Ben, Pei,
How does your system handle choices such as
--- Lukasz Stafiniak [EMAIL PROTECTED] wrote:
Iddo Lev has a more practical answer:
http://www.stanford.edu/~iddolev/pulc/current_work.html
Just looking at it briefly, it appears to clearly present the many problems
with natural language understanding (i.e. various forms of ambiguity). Then
On 5/8/07, James Ratcliff [EMAIL PROTECTED] wrote:
Pei,
The only problem I see with choosing the first one Pei, is that given the
3 choices, and taking #1, if the system does not learn anything extra that
would help it make a decision, it would be forever stuck in that loop, and
never able to
Are there any projects that allow people to help?
__
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
-
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change
I really hate to get into this endless discussion. I think everyone agrees
that some randomness in AGI decision making is good (e.g. learning through
exploration). Also it does not matter if the source of randomness is a true
random source, such as thermal noise in neurons, or a deterministic
This work is a progress along a clearly stated line of research. It is
about ways of managing ambiguity. The dissertation is not about a
complete AGI system, it does not go into machine learning. It is not
about discovering meaning, but analysing meaning: the interpretation
of output is fully
Personally, I found the dissertation highly enlightening and helpful.
Yes, it addresses ambiguity problems with rules (though I debate both of
Matt's descriptors -- the term huge and the term complicated) without
specifying how these rules might be machine-learned -- but doing so is still
a
You are welcome. Indeed, I was tempted to keep it for myself ;-)
As for learning rules, I guess you know the work
http://citeseer.ist.psu.edu/605753.html or similar. In practical
contexts, it must be integrated with learning the semantical lexicon
(e.g., feature structures), and thus, the
You are welcome. Indeed, I was tempted to keep it for myself ;-)
Please. DON'T!:-)
As for learning rules, I guess you know the work
http://citeseer.ist.psu.edu/605753.html or similar. In practical
contexts, it must be integrated with learning the semantical lexicon
(e.g., feature
19 matches
Mail list logo
