Hi Stephen, nice to meet you. When I search the web for critiques
of CYC, I can only find stuff from '90-95. If no one has written
critiques of CYC since then, perhaps you could comment on how applicable
those early critiques would be to the current system.
For example, would CYC today at least
2008/11/30 Ben Goertzel [EMAIL PROTECTED]:
Could you give me a little more detail about your thoughts on this?
Do you think the problem of increasing uncomputableness of complicated
complexity is the common thread found in all of the interesting,
useful but unscalable methods of AI?
Jim
Hi,
I have proposed a problem domain called function predictor whose
purpose is to allow an AI to learn across problem sub-domains,
carrying its learning from one domain to another. (See
http://www.includipedia.com/wiki/User:Cabalamat/Function_predictor )
I also think it would be useful if
A little more poking around reveals further evidence that supports the
glocal model of brain memory (they talk about a
distributed plus hub model, which is part of the glocality idea,
though missing the nonlinear-attractor aspect that I think is critical
to distributed memory)
2008/11/29 Matt Mahoney [EMAIL PROTECTED]:
The general problem of detecting overfitting is not computable. The principle
according to Occam's Razor, formalized and proven by Hutter's AIXI model, is
to choose the shortest program (simplest hypothesis) that generates the data.
Overfitting is
AIXI is a purely theoretic construct, requiring infinite computational resources
AIXItl is a version that could be implemented in principle, but not in
practice due to truly insane computational resource requirements
Whether the line of thinking and body of theory underlying these
things can be
2008/11/30 Ben Goertzel [EMAIL PROTECTED]:
Hi,
I have proposed a problem domain called function predictor whose
purpose is to allow an AI to learn across problem sub-domains,
carrying its learning from one domain to another. (See
Robin,
While I was at Cycorp, a concerted effort was made to address Vaughan Pratt's
test questions. I recall that most of them required the addition of facts and
rules into the Cyc KB so that they would answer. I believe that a substantial
portion are included in the Cyc query regression
Ben,
Cycorp participated in the DARPA Transfer Learning project, as a subcontractor.
My project role was simply a team member and I did not attend any PI
meetings. But I did work on getting a Quake III Arena environment working at
Cycorp which was to be a transfer learning testbed. I also
Matt Taylor was also an intern at Cycorp where was on Cycorp's Transfer
Learning team with me.
-Steve
Stephen L. Reed
Artificial Intelligence Researcher
http://texai.org/blog
http://texai.org
3008 Oak Crest Ave.
Austin, Texas, USA 78704
512.791.7860
From:
Regarding winning a DARPA contract, I believe that teaming with an
established contractor, e.g. SAIC, SRI, is beneficial.
Cheers,
-Steve
Yeah, I've tried that approach too ...
As it happens, I've had significant more success getting funding from
various other government agencies ... but
Stephen,
Does that mean what you did at Cycorp on transfer learning is similar
to what Taylor presented to AGI-08?
Pei
On Sun, Nov 30, 2008 at 1:01 PM, Stephen Reed [EMAIL PROTECTED] wrote:
Matt Taylor was also an intern at Cycorp where was on Cycorp's Transfer
Learning team with me.
-Steve
Philip Hunt wrote:
2008/11/29 Matt Mahoney [EMAIL PROTECTED]:
The general problem of detecting overfitting is not computable. The
principle according to Occam's Razor, formalized and proven by
Hutter's AIXI model, is to choose the shortest program (simplest
hypothesis) that generates the data.
Pei,
Matt Taylor's work at Cycorp was not closely related to his published work at
AGI-08.
Matt contributed to a variety of other Transfer Learning tasks, and I cannot
recall exactly what those were.
-Steve
Stephen L. Reed
Artificial Intelligence Researcher
http://texai.org/blog
Ed,
Unfortunately to reply to your message in detail would absorb a lot of
time, because there are two issues mixed up
1) you don't know much about computability theory, and educating you
on it would take a lot of time (and is not best done on an email list)
2) I may not have expressed some of
Ed,
I think that we must rely on large collections of relatively simple
patterns that are somehow capable of being mixed and used in
interactions with the others. These interacting patterns (to use your
term) would have extensive variations to make them flexible and useful
with other patterns.
But quantum theory does appear to be directly related to limits of the
computations of physical reality. The uncertainty theory and the
quantization of quantum states are limitations on what can be computed by
physical reality.
Not really. They're limitations on what measurements of
On Mon, Dec 1, 2008 at 11:19 AM, Ed Porter [EMAIL PROTECTED] wrote:
You said QUANTUM THEORY REALLY HAS NOTHING DIRECTLY TO DO WITH
UNCOMPUTABILITY.
Please don't quote people using this style, it hurts my eyes.
But quantum theory does appear to be directly related to limits of the
Regarding the uncertainty principal, Wikipedia says:
In quantum physics, the Heisenberg uncertainty principle states that the
values of certain pairs of conjugate variables (position and momentum, for
instance) cannot both be known with arbitrary precision. That is, the more
precisely one
HI,
In quantum physics, the Heisenberg uncertainty principle states that the
values of certain pairs of conjugate variables (position and momentum, for
instance) cannot both be known with arbitrary precision. That is, the more
precisely one variable is known, the less precisely the other is
OTOH, there is no possible real-world test to distinguish a true
random sequence from a high-algorithmic-information quasi-random
sequence
So I don't find this argument very convincing...
On Sun, Nov 30, 2008 at 10:42 PM, Hector Zenil [EMAIL PROTECTED] wrote:
On Mon, Dec 1, 2008 at 3:09 AM,
On Mon, Dec 1, 2008 at 3:09 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
But quantum theory does appear to be directly related to limits of the
computations of physical reality. The uncertainty theory and the
quantization of quantum states are limitations on what can be computed by
physical
On Mon, Dec 1, 2008 at 4:44 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
OTOH, there is no possible real-world test to distinguish a true
random sequence from a high-algorithmic-information quasi-random
sequence
I know, but the point is not whether we can distinguish it, but that
quantum
On Mon, Dec 1, 2008 at 4:53 AM, Hector Zenil [EMAIL PROTECTED] wrote:
On Mon, Dec 1, 2008 at 4:44 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
OTOH, there is no possible real-world test to distinguish a true
random sequence from a high-algorithmic-information quasi-random
sequence
I know,
But I don't get your point at all, because the whole idea of
nondeterministic randomness has nothing to do with physical
reality... true random numbers are uncomputable entities which can
never be existed, and any finite series of observations can be modeled
equally well as the first N bits of an
On Mon, Dec 1, 2008 at 4:55 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
But I don't get your point at all, because the whole idea of
nondeterministic randomness has nothing to do with physical
reality...
It has all to do when it is about quantum mechanics. Quantum mechanics
is non-deterministic
On Mon, Dec 1, 2008 at 4:55 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
But I don't get your point at all, because the whole idea of
nondeterministic randomness has nothing to do with physical
reality...
I don't get it. You don't think that quantum mechanics is part of our
physical reality (if
On Sun, Nov 30, 2008 at 11:48 PM, Hector Zenil [EMAIL PROTECTED] wrote:
On Mon, Dec 1, 2008 at 4:55 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
But I don't get your point at all, because the whole idea of
nondeterministic randomness has nothing to do with physical
reality...
I don't get it.
On Mon, Dec 1, 2008 at 6:20 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
On Sun, Nov 30, 2008 at 11:48 PM, Hector Zenil [EMAIL PROTECTED] wrote:
On Mon, Dec 1, 2008 at 4:55 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
But I don't get your point at all, because the whole idea of
nondeterministic
On Nov 30, 2008, at 7:31 AM, Philip Hunt wrote:
2008/11/30 Ben Goertzel [EMAIL PROTECTED]:
In general, the standard AI methods can't handle pattern recognition
problems requiring finding complex interdependencies among multiple
variables that are obscured among scads of other variables
Hector Zenil wrote:
On Mon, Dec 1, 2008 at 6:20 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
On Sun, Nov 30, 2008 at 11:48 PM, Hector Zenil [EMAIL PROTECTED] wrote:
On Mon, Dec 1, 2008 at 4:55 AM, Ben Goertzel [EMAIL PROTECTED] wrote:
But I don't get your point at all, because the
31 matches
Mail list logo
