can system dynamics learn
Posted: Mon Apr 21, 2003 8:46 pm
Hi everybody
I have questions but no answers.
First about the question and the object of the learning:
What is the object of the learning?
Is the model learning about the way it as a model is working
or is it learning something about the subject it is supposed to
work on?
Is the learning inside or outside the boundaries of the SD model?
If it is inside, is it inside the strict conception of the model or
can it discover something outside of that conception? This looks
difficult to imagine because even inside the boundaries of the model
the learning will be limited by the very conception of the model.
Is the model able to extend its original boundaries?
This looks like science fiction.
Is the model able to learn by itself ot does it need the help of a human?
In the case of inside learning or outside learning or both?
Who is then learning the model or the human?
Is there not a dynamic system, LIFE an exemple of a system having
learned outside or its original boundaries?
But was it learning or the selection of the most fit through hazard?
Is then pure learning outside or its original boundaries, only the
effect of chance?
Would a well conceived dynamic system that has boundaries
large enough for the subject studied, FREE to make stochastic experiences
even out of the bound ot its conception, having plenty of time and a very small
chance but still positive to find something a possible candidate?
Would "he"?
J.J. Laublé.
From: "ALLOCAR SRASBOURG" <allocar-strasbourg@wanadoo.fr>
I have questions but no answers.
First about the question and the object of the learning:
What is the object of the learning?
Is the model learning about the way it as a model is working
or is it learning something about the subject it is supposed to
work on?
Is the learning inside or outside the boundaries of the SD model?
If it is inside, is it inside the strict conception of the model or
can it discover something outside of that conception? This looks
difficult to imagine because even inside the boundaries of the model
the learning will be limited by the very conception of the model.
Is the model able to extend its original boundaries?
This looks like science fiction.
Is the model able to learn by itself ot does it need the help of a human?
In the case of inside learning or outside learning or both?
Who is then learning the model or the human?
Is there not a dynamic system, LIFE an exemple of a system having
learned outside or its original boundaries?
But was it learning or the selection of the most fit through hazard?
Is then pure learning outside or its original boundaries, only the
effect of chance?
Would a well conceived dynamic system that has boundaries
large enough for the subject studied, FREE to make stochastic experiences
even out of the bound ot its conception, having plenty of time and a very small
chance but still positive to find something a possible candidate?
Would "he"?
J.J. Laublé.
From: "ALLOCAR SRASBOURG" <allocar-strasbourg@wanadoo.fr>