Re: topmind: ID is potentially testable


Zachriel wrote:
"topmind" <topmind@xxxxxxxxxxxxxxxx> wrote in message
news:1146637355.405987.296460@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Zachriel wrote:
"topmind" <topmind@xxxxxxxxxxxxxxxx> wrote in message
news:1146552817.919692.59850@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

Zachriel wrote:
"topmind" <topmind@xxxxxxxxxxxxxxxx> wrote in message
news:1146460600.803368.148690@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Zachriel wrote:
"topmind" <topmind@xxxxxxxxxxxxxxxx> wrote in message
news:1146377988.818732.233220@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

<snip>


If there was a bit-map of the Mona
Lisa within the human genome, there
is a very high probability it would have
been noticed because it would stick
out like a statistical sore thumb.

(Note the qualification to probability and to the specific encoding
scheme.)

Not all encryptions of an image would be detectable. There are
unlimited
numbers of methods of encoding and encryption, including
encryptions
unbreakable with current technology.


I realize that, but that is not the issue being raised here. Most
of
the algorithms listed by those guys appear to look for repeating
codon
sequences, perhaps with typical transpositions and mutuations as
found
in biology. However, those kinds of patterns would NOT be found in
typical bitmapped images (such as Mona).


"Most". What about the rest?


Perhaps all. If you have a favorite that WOULD HAVE detected Mona if in
those studies, by all means point the damned thing out. What are you
afraid of? Scrutity?



Generalized pattern recognition software has been used repeatedly by
genomic
researchers. Lilith provided you ample direction to understanding. But
it's
hard to see when you cover your eyes.


Other than the human mind, THERE IS NO SUCH THING as "Generalized
pattern recognition software".


Did I point out that the genome has been printed out in a variety of
formats
(sometimes in color) and people have visually scanned the pages?


Unless the scan lines are lined up properly (the line width is known),
most bitmapped images will not stand out from randomness to the human
eye, especially if the palette mapping is not known (which is the
likely case).


So, in other words, the human mind is not as generalized as you had
indicated.

I never claimed it was perfect.



That is why it is best to use a computer to try the
myriad scan-line widths that would need to be tested.

The human eye would be best left to evaluating the candidates
identified by automated processes, not the other way around.


You want to detect an image, specifically an image of the Mona Lisa, but you
don't even know what it means to be an image.

I already gave several suggestions. But actually this is moot because
one can train neural nets or genetic algorithms (evolution!) to find
them based on training sets. One does not have to "define" the training
set, only provide examples. (Different alignments (framing) will
probably still have to be tried.)


More importantly, you don't
even care.


This has nothing to do with the price of beans.





Did you know that the human mind is not completely generalized, but works
within a specific sensory and analytical framework?

I am not sure what your point is here.


Of course not.


Grump.





This is part of the reason why humans
flip burgers for 7 dollars an hour instead of machines for 50-cents per
hour.

Yes, there are algorithms that detect snippets of exact matches or
nearly exact matches (codon sequences), but that is not what you need
for most images. I've studied the bitmaps of various images.


Tell, us. Where is your statistical analysis of the content of a bit-map
of
the Mona Lisa? Of generalized images?

Why is it my burden to show you? Youses are the ones who claimed that
stuff already tried would have found Mona (as a test case). You claim,
your prove.


You just claimed that you had "studied the bitmaps of various images". Was
this a lie?

topmind: "I've studied the bitmaps of various images."


I never claimed I wrote image detection algorithms.



"Good enough for Mona" is NOT the default.


Moving the goalposts and consistently ignoring evidence that is contrary to
your prejudice.

Yeah yeah, I'm all bad and evil and manipulative and you an angel, blah
blah blah. It's getting old. I already know you hate my guts and think
I am Hitler reincarnated. Repetition is unnecessary.


Recurrence time statistics: versatile tools for genomic DNA sequence
analysis.
Cao Y, Tung WW, Gao JB, Qi Y.

"Our method ... to extract all the repeat-related and periodic or
quasi-periodic features ... without any prior knowledge on the consensus
sequence of those features"
http://tinyurl.com/gs8vn

I suspect that "consensus" is the "feedback" talked about in a nearby
reply to another. It is not the training set itself that is being done
"without".



--

topmind: "Whether my victories are real or hallucinations, they are
satisfying either way."

I don't see your point in repeating this quote. Then again, a grump
like you is probably against any satisfaction, real or imagined.


Zachriel
http://zachriel.blogspot.com/


-T-

.

Relevant Pages

  • Re: topmind: ID is potentially testable
    ... Just show some evidence that outliers or oddities have been ... that bitmaps create a statistical signature that does not exist in the ... Perhaps I should submit Finding Mona to the Journal Nature! ... algorithms would have to be 'tuned' to find such images, ...
    (talk.origins)
  • Re: topmind: ID is potentially testable
    ... typical bitmapped images (such as Mona). ... You just claimed that you had "studied the bitmaps of various images". ... Recurrence time statistics: versatile tools for genomic DNA sequence ...
    (talk.origins)
  • Re: topmind: ID is potentially testable
    ... Lisa within the human genome, ... typical bitmapped images (such as Mona). ... Generalized pattern recognition software has been used repeatedly by ...
    (talk.origins)
  • Re: topmind: ID is potentially testable
    ... If there was a bit-map of the Mona ... statistical measures of the bitmap and genome would have indicate. ... easily said to contain the Mona Lisa, just as the digits of PI could be. ... images is different than those found in typical biological processes. ...
    (talk.origins)
  • Re: topmind: ID is potentially testable
    ... If there was a bit-map of the Mona ... I've taken courses in pattern recognition. ... images is different than those found in typical biological processes. ... C - Dark Grey ...
    (talk.origins)