Bill's Gate III

From Mervyn van Kuyen (970926 20:30 CET)

Bill, please allow me another chance to make my point.

When you wrote...

[Bill Powers (970921.0728 MDT)]

the nature of neural signals says they can't go negative, so if you have
an excitatory reference signal and an inhibitory effect of the perceptual
signal at the comparator, a zero reference signal guarantees zero error
for all magnitudes of perceptual signal.

... and later on about that 'inverted comparator' it struck me that you
left out the third member of this class of 'comparators' (although I
wouldn't call the 'inverted comparator' a comparator), which I believe is
at least as interesting as the other two:

      (Fig.1) (Fig.2) (Fig.3)

      NETWORK NETWORK NETWORK
         A A A
         > > >
    +->(EXC) +->(EXC)<-+ (EXC)<-+
    > A | A | A |
    > > > > > > >
    > (INH)<-+ +->(INH)<-+ +->(INH) |
    > 1 | | 2 | | 1 |
    > > > > > >
   REF SENS REF SENS REF SENS

It is interesting to see how these members each have their distinctions:

(Fig.1) This is the fundament of PCT.
o Active reference combined with no input triggers the network.
This gate only compares the reference and sensory input only as far as it
actively tests. This allows the network to be unaware and undisturbed of
its incompleteness. This fruit is picked by PCT. This also implies that
there must be a predefined reference to get the system started. This
demand is met by PCT.

(Fig.2) This is the fundament of my Neural Servo.
o Any mismatch between reference and input triggers the network.
The fact that input (without reference) is able to trigger the network all
by itself allows the network start from scratch in its development. This
fruit is picked by a Neural Servo. The extra feedback generated by
feeding transformed mismatch back as reference and by allowing the input
to trigger the network (in the absence of reference) makes this system
violently dynamical. This demand is met by the Neural Servo's continuous
modulation, which can be suggested to explain the workings of EEG waves.

(Fig.3) This is the fundament of spotlight, s-r and filter models
o Reference signals block the input
This configuration is nothing but a tool for intelligent systems, it does
not generate any information about the relation between input and
reference, it is a tool for disposing information. This implies that this
configuration is not a comparator of any sort. This doesn't give us much
hope for these models :wink:

The only point is was trying to make is that there is a fundamentally
different comparator available, something you seem to be unaware of. And
although you are right about the nature of signals ("they can't go
negative"), you shouldn't imply that (Fig.1) is the closest thing to
subtraction (in order to obtain the difference between two signal in a
servo-like manner): Fig.2 can be described by |ref-sens| ! Maybe, with
some disciplined thinking, we can work out yours? :wink:

Regards, Mervyn

[From Bill Powers (970928.0504 MDT)]

Mervyn van Kuyen (970926 20:30 CET) --

Bill, please allow me another chance to make my point.

OK, but you're going to have to give me some help.

     (Fig.1) (Fig.2) (Fig.3)

     NETWORK NETWORK NETWORK
        A A A
        > > >
   +->(EXC) +->(EXC)<-+ (EXC)<-+
   > A | A | A |
   > > > > > > >
   > (INH)<-+ +->(INH)<-+ +->(INH) |
   > 1 | | 2 | | 1 |
   > > > > > >
  REF SENS REF SENS REF SENS

It is interesting to see how these members each have their distinctions:

Before you try to explain that, I need to know what the above arrangement
is designed to accomplish, whichever version you prefer. What does the
network, taken as a unit, _do_? And to what aspect of the behavior of a
living system does its behavior correspond? Does the network have any
physical effects on its environment? Does the sensor detect something in
the physical environment? And what is it that generates REF?

Best,

Bill P.