Re: What is conditioning?


JGCASEY wrote:
On May 8, 12:52 am, Wolf <ElLoboVi...@xxxxxxxxxx> wrote:
[...]
Wolf:
A sensor filters

....

Wolf:
All sensors are filter by definition. Otherwise they
wouldn't be sensors.

JC:
Your words were "a sensor filters" not "sensors filter"
but I get what mean now.


JC:
I view a filter more like this:

input output1 output2 output3
HDSUECXEW H C w
EOCOIUEWC CC W
EWOWIECWQ C WWW
BCMXEWOCW CC WW
OCWKTPGUY C

Wolf:
So you have three sensors here: one responds only to
H, one only to C, one only to W.


JC:
Actually they are not sensors they are circuits that
compute the existence of H in one case C in another.
For example let us say you have these areas.

*** ** *******
*** * ******
*** ** *******

How would you passively filter them into three groups
according to size? Can't exactly drop them though a
series of passive wire meshes but you can count the
number of *'s and then sort using if then statements
or their neural equivalent. This is active filtering.
The outputs can then be associated with each other
or other things.
========

JC:
For example in a visual system that filters out
high contrast traffic signs you might start with
a local threshold filter. Then next set of filters
might filter out the resulting areas into sizes
and shapes. The resulting outputs can become
"symbols" to be manipulated at a higher level.

Wolf:
It seems to me you are describing not sensors but
data processing.

JC:
I am describing "filtering". You know like a sand
filter can separate particles of different sizes.
A sensor like a wire mesh can be used to filter
but the filtering process can use active components
as well as passive components.


Wolf:
Of course, a sensor array may be structured so that
its structure processes the data. This is in fact
how the retina works. Its three layers of neurons
are interconnected such that edges, for example,
are detected in the retina.


JC:
A sensor in the eye, like a sensor in a robot, I think
is meant to mean the device that converts energy from
one form to the electrical energy used in the brain.
After that the energy in or out is electrical. You are
not filtering out different kinds of energy rather you
are filtering different kinds of invariant patterns.
========

JC:
In general a filtering process extracts invariants
from an ever changing input. At the input side you
have changing pixel values (thinking computers here)
as say a cow moves about in front of the camera.

At the output side you have the cow symbol which
remains constant even though the input side consists
of ever changing patterns of pixel values as the
cow (or camera, or other objects in view) move
about.


Wolf:
IMO the visual system doesn't output a symbol, it
outputs a shape. Just how it does this isn't easy
to see. ;-)

The language system receives some output from the
visual system and outputs a speech act, which we
abstract as a symbol. But abstracting a symbol
from this furshlugginer* mess is IMO the task of
neither the language nor the visual systems. I
speculate another system. Maybe located in the
prefrontal cortex.

*well worn, beat up

JC:
Call it a symbol, shape or maybe some other kind of
invariant pattern that can associate with other
invariant patterns whatever form it and they take.
======


Wolf:
I didn't think building a machine from mechanical or math
principles is going to do it. Unless of course you are
aiming at a math machine.

JC:
May I direct your attention to a little book I have
mentioned before: "Life's Other Secret, the new
mathematics of the living world" by Ian Stewart. It
is written for the layman and is a very easy read and
shows the role mathematics plays in understanding the
behavior and evolution of biological machines.
=======

Wolf:
IMO you won't get a learning machine if you write code.
You'll get a simulation of a learning machine. That will
be good enough for many if not most practical purposes,
eg, a smart spell checker, that learns my characteristic
mistakes and corrects them as I type.

JC:
Writing code is simply a way to wire up the components
in a general purpose computer to make it a special
purpose machine.
========

JC:
You cannot predict with 100% the response of a program
for any given environment without knowing the internal
details of the program.

Wolf:
I don't think you can predict it even then. For if you
could do that, you could write bug free programs.

JC:
What makes you think there aren't any bug free programs :)

If someone gives you the algorithm used to compute the
square root of a number I think you could say with some
certainty its range of behaviors. However if all you
has was a few input/output examples you could never be
sure what the range might be.
=======


Wolf:
Exercise for the interested student: Just what is the
output of a Mandelbrot program? And why is it conceptually
wrong to equate that with its behaviour?

JC:
So the program isn't behaving? Or it isn't a good analogy
with an animals behavior? It was not meant as an analogy
with an animals behavior only with the difficulty in
reverse engineering from behavior. If the behavior is
simple like Microsoft Windows you might be able to duplicate
the behavior using completely different code.
=======


Wolf:
(Actually, I don't think it's sensible to talk about the
behaviour of a program.)

JC:
Don't confuse the program with the machine it creates by
"wiring up" the components in a general purpose computer.
You could skip the programming bit and just wire it up
yourself. AI is about the behavior of machines.

A lot of people seem to forget that a computer is just
a collection of parts that have to be wired up. They then
often talk about what computers can or cannot do not
realising they are talking about what programs have or
have not done.
==========


Wolf:
Sideways jump to illustrate a point: a calculator does
arithmetic. The kid that uses the calculator does math.
Why is a calculator such a simple machine? And why don't
we have math machines yet?

JC:
By math machine do you mean a machine that does math
the way the kid does?

And what do you think most students do? I think most
learn a set of rules for solving a problem. You might
for example give them a problem in English and they
have to translate it into a algebraic form to solve
according to the rules they have been programmed with.

(see Bobrow's STUDENT program which solves high-school
algebra problems stated in English).

And why do you think we haven't got a math machine yet?



--
JC




.

Relevant Pages

  • Re: blood pressure!
    ... WIth the filter i surely agree have lot of work needs to be done! ... power is hooked up backwards, if you've got everything miswired, it's ... and mount it right next to the sensor. ... if you want and apply that pressure again, or you can just ground one ...
    (sci.electronics.basics)
  • Re: Cassini: where are colors and detail?
    ... While one filter response may be dropping as you change ... Algorithms in consumer cameras for interpreting Bayer sensors also go ... have to do spectrometry using a Bayer filtered CCD as the sensor. ...
    (sci.astro.amateur)
  • Re: integration of a continuous function
    ... Is he not using a piezoelectric sensor, ... Without motion there is no voltage. ... filter that has a frequency response similar to the frequency response of ... response of summing the samples (moving average filter) is actually 1/sinnot ...
    (comp.dsp)
  • =?iso-8859-1?q?Re:_Leica_M8_-_L=F6sung_f=FCr_Probleme_angek=FCndigt?=
    ... das mit der Beschränkung auf 2 Filter war der 3. ... diesen Fehler nun logistisch abwickeln will, muss Leica selbst wissen. ... wieviel man in Sandwichbauweise auf den Sensor packt an Mikrolinsen, ... ist das IR zu filtern ist meines Erachtens eben nur eine Behauptung ...
    (de.rec.fotografie)
Loading