In principle video on the internet may become a useful resource for
AGI learning. However in practice the previous comments hand wave
over a lot of very complex details, in a similar manner to how some of
the early AI pioneers believed that visual interpretation would be
easy.
To be able to
Well yes, breezing over the details, there are many, and building the
software would be a full time effort for many months at least. I'm sure
there are some commercial products already built and they would be expensive
to license into an AGI application. It would be nice if the solution would
be
Bob Mottram wrote:
it seems infeasible that 2D templates
need to be created for every possible viewing angle and scale of an
object
I think this is similar to how our vision works. We have visual short
term memory that seem to hold 2D templates for a few seconds.
We have specialized
Self, POV, I - these may be more biological, evolutionary optimized,
physical things not required in an AGI. Picture two fused human brains
created in a lab 50 years from now by some mad scientist. What type of self
would this be? In AGI software self could be geographically distributed
A think that a subset of visual imagery, the most commonly observed
objects, may be stored in terms of 2D templates but I doubt that most
imagery is represented in this way. Our brains are fundamentally
designed to deal with 3D or 4D information. We don't live in a
flatland. A greater level of
On 8/4/07, John G. Rose [EMAIL PROTECTED] wrote:
Why do you say that should we train ourselves to prefer solutions that use
more computing power? Is this because those solutions in general are more
whole and manageable verses tweaked, hacked and optimized with loss of
adaptability?
Yes, and
I'm still v. much groping here - still trying to absorb the idea of a self
as essential to vision and visual object recognition.
I suspect though that the kind of ideas you're suggesting - of some AGI
without a self - are strictly a fiction and illusion.
The self, it seems to me, is
Robert has some good, sound, fundamental points here.
Rand seems to not understand that a psychopath in respect to humanity
has no value, that it is a mutation that occurs, but in a social
community it has no value, in the sense that human values have to be
both selfish and altruistic to have
What you see is dependent on your reaction.
How you react is dependent on what you see.
Memory recall is an reaction. You are reacting to the image by recalling
things relating to the image.
Reaction is impossible if and only if you didn't see it.
That means that not reacting to a stimuli is
For object recognition think of OCR. OCR is very good now. It's not AGI
it's, I dare to say, basically a mechanical operation. For general object
recognition the same type of speed and accuracy would fit the bill. Is
there a self needed for OCR?
For us yes a self is integral. If you are
9p9p
ASEORIA LEGAL
¿NECESITA ABOGADO POR PROBLEMAS INQUILINARIOS, DEUDAS POR COBRAR, DIVORCIOS,
HERENCIAS, CUMPLIMIENTOS O RESCICIONES DE CONTARTOS CIVILES O MERCANTILES?
CONTAMOS CON MAS DE 20 AÑOS DE EXPERIENCIA PROFESIONAL
LLÁMENOS AL BUFFETE JURÍDICO PÉREZ GAVILAN Y SI REFIERE QUE
11 matches
Mail list logo
