Emergent Behavior
Philipp A. Djang
I have seen and read, EB is an intuitive concept, defined from the eye of
the beholder.
For example, flocking is an example of EB. Can someone point me to a
mathematical definition of flocking? Should EB follow some set of universal
or local principles? Is it possible to create definitions of EB based on
local logic, formal concept analysis or category theory?
Thanks,
Philipp
David C. Ullrich
<pdj...@psl.nmsu.edu> wrote:
>I am looking for a rigorous definition of emergent behavior (EB).
Good luck.
> From what
>I have seen and read, EB is an intuitive concept, defined from the eye of
>the beholder.
That's the way it seems to me as well. The "definition" seems to
be more or less that a behavior is "emergent" when it could not
have been predicted from the behavior of the pieces of the
system. But in fact the actual behavior _is_ predictable, we're
just not able to do the calculations.
Robert J. Kolker
"Philipp A. Djang" wrote:
> I am looking for a rigorous definition of emergent behavior (EB). From what
> I have seen and read, EB is an intuitive concept, defined from the eye of
> the beholder.
Emergence is what happens when Reductionism hits the wall.
Bob Kolker
Jorn Barger
> what I have seen and read, EB is an intuitive concept, defined from the
> eye of the beholder.
I've never thought about this before, but surely it must be grounded in
human psychology?
There are 'starter concepts' that are easy to explain unambiguously (eg
the rules of Conway's Life), but if we already understood them
completely, nothing new could possibly emerge.
Because our mathematical intuitions evolved by our ancestors'
interactions with the physical world, we have fragmentary working models
of numbers and shapes, etc, but can only work out their full
implications slowly, by 'longhand'.
--
http://www.robotwisdom.com/ "Relentlessly intelligent
yet playful, polymathic in scope of interests, minimalist
but user-friendly design." --Milwaukee Journal-Sentinel
Randall R Schulz
The putative definition for "emergence" you quote from
<http://www.calresco.org/sos/sosfaq.htm> seems to me to be rather weak.
This definition's use of the phrase "higher level" seems problematic,
too, since it appears that much of the definition's substance rests on
this phrase, yet that phrase does not seem very meaningful in itself.
This definition would admit, for example, that the emergent property of
"boxness" arises from the bringing together of six suitably sized
rectangular sheets of a rigid planar material arranged in the customary
box structure. A hinged lid would probably be required, too...
Perhaps that's desirable. Maybe the notion should extend to the trivial
(a box) as well as the profound (human cognition and consciousness).
Perhaps.
However, I'm afraid that it can be seen to apply to absolutely
everything that has a composite nature, from the nucleon (proton,
neutron, antiproton--three quarks bound by the exchange of strong force
gluons) up to the universe as as whole. It could probably even be said
to apply to the genuinely fundamental particles such as the leptons
(electrons, e.g.) or even force mediating particles (photons, e.g.).
Actually, about the only thing that this definition probably omits is
"the universe as a whole," since the universe has no outside from which
to observe its "emergent" behavior.
To me, this over-broadness undermines its value as a definition.
Randall Schulz
Mountain View, CA USA
Martin Sewell wrote:
>"Philipp A. Djang" <pdj...@psl.nmsu.edu> wrote:
>
>>I am looking for a rigorous definition of emergent behavior (EB). From what
>>I have seen and read, EB is an intuitive concept, defined from the eye of
>>the beholder.
>>
>
>"2.3 What is emergence ?
>The appearance of a property or feature not previously observed as a
>functional characteristic of the system. Generally, higher level
>properties are regarded as emergent. A automobile is an emergent
>property of its interconnected parts. That property disappears if the
>parts are disassembled and just placed in a heap."
>
>
>Cheers
>
>Martin
>
Luke Kaven
the others sciences) has a historical basis in the "emergentism" of the
early-mid 20th c. It is commonly used as a casual matter among scientists.
When scientists try to formalize the concept, they find it elusive. It is
understandably so, since the concept turns upon a number of long-standing
problems in metaphysics and epistemology, and that kind of training
typically falls outside contemporary scientific education. So the term
"emergent" is typically used by scientists as a place holder, a kind of
necessary gloss over a problem that lays (understandably) outside their area
of knowledge.
Here is what I think is most pertinent:
The theoretical status of "emergence" is tied up in the debates pertaining
to reductive versus non-reductive materialism, as well as to the complex
related forms of "soft" (non-Cartesian) dualism (eg., supervenience). The
problem of how high-level things related to low-level things of course has
been the subject of an industry in philosophy for hundreds of years. The
problem of mind and body is just a particularly telling case of the general
problem.
The term "emergence" was employed in the early-mid 20th c. to designate one
particular approach to the problem (C.D. Broad, I believe). For a good
criticism of Emergentism, and a discussion of supervenience, see Jaegwon Kim
(in _Supervenience and Mind_, Cambridge University Press). For a very good
survey of different approaches, see the readings in _Mental Causation_ (Heil
and Mele (eds.), Oxford University Press). As I suggested, I don't believe
that scientists use the term "emergent" with much univocity. It is just
used as a stand-in term that very smartly identifies the place where
something very much is needed!
To get one highly influential approach to the subject of behavior, I would
recommend seeing Fred Dretske's _Explaining Behavior_ (MIT Press). Its a
rather short book, but one of stunning insight, and it has become a classic
in philosophy. The reasons-based account of behavior I think stands a
chance to do the kind of work that scientists can use, if they can bridge
the intertheoretical gap.
Best of luck, Luke Kaven
Jorn Barger <jo...@enteract.com> wrote in message
news:1f1kv35.1m7wqgq1trllnlN%jo...@enteract.com...
Luke Kaven
"emergence" prematurely. In another post, I noted that the question of what
scientists are referring to in their uses of "emergence" is tied up in a
debate in metaphysics and epistemology of long-standing. This particular
definition does not offer us anything we can avail ourselves of.
A quick critique of this one:
1) The question of whether a particular property has been observed or is
observable is immaterial.
2) An automobile is not a property!
It will keep coming back to the question of which brand of reductive or
non-reductive materialism one accepts. For me, its teleosemantics.
Luke
Martin Sewell <mar...@martinsewell.com> wrote in message
news:d5n3tt8qnsujjmk8s...@4ax.com...
> On Fri, 19 Oct 2001 15:33:22 -0600, "Philipp A. Djang"
> <> wrote:
>
> >I am looking for a rigorous definition of emergent behavior (EB). From
what
> >I have seen and read, EB is an intuitive concept, defined from the eye of
> >the beholder.
>
> "2.3 What is emergence ?
> The appearance of a property or feature not previously observed as a
> functional characteristic of the system. Generally, higher level
> properties are regarded as emergent. A automobile is an emergent
> property of its interconnected parts. That property disappears if the
> parts are disassembled and just placed in a heap."
>
Luke Kaven
Randall R Schulz <rrsc...@cris.com> wrote in message
news:3BD1EF5B...@cris.com...
> Martin,
>
> The putative definition for "emergence" you quote from
> <http://www.calresco.org/sos/sosfaq.htm> seems to me to be rather weak.
> Perhaps that's desirable. Maybe the notion should extend to the trivial
> (a box) as well as the profound (human cognition and consciousness).
> Perhaps.
That has always been inherent in the project of the mind-body problem. The
mind is just a particularly telling case.
> However, I'm afraid that it can be seen to apply to absolutely
> everything that has a composite nature, from the nucleon (proton,
> neutron, antiproton--three quarks bound by the exchange of strong force
> gluons) up to the universe as as whole. It could probably even be said
> to apply to the genuinely fundamental particles such as the leptons
> (electrons, e.g.) or even force mediating particles (photons, e.g.).
> Actually, about the only thing that this definition probably omits is
> "the universe as a whole," since the universe has no outside from which
> to observe its "emergent" behavior.
Observability aside. Some do think that teleological characterizations can
be given to even low-level phenomena (Lycan thinks that molecules are
"molecule-ers", for example). This one approach to the mind-body problem
has some interesting implications for a theory of nature.
> To me, this over-broadness undermines its value as a definition.
Though it wasn't an unintelligent guess, it would get at best a polite laugh
in any philosophy seminar. .
Luke
Jorn Barger
> The popular usage of the term "emergent" in computer science (and some of
> the others sciences) has a historical basis in the "emergentism" of the
> early-mid 20th c.
The difference is that the CS people *know* what they're talking
about... literally. They're talking about programs that re-channel
electron flows among microchips, according to simple rules of boolean
logic.
So when behavior appears to 'emerge' in this realm, there's no cause for
metaphysical onanism...
[blah blah blah]
> Though it wasn't an unintelligent guess, it would get at best a polite
> laugh in any philosophy seminar.
Yeah, like that.
Lionel Barnett
> I am looking for a rigorous definition of emergent behavior (EB). From
> what I have seen and read, EB is an intuitive concept, defined from the
> eye of the beholder.
A working definition of "emergent behaviour" (in the alife context) is:
"something complicated happened that I don't understand and can't be arsed
to describe in more precise and meaningful language".
--
Lionel Barnett
CALResCo
>The putative definition for "emergence" you quote from
><http://www.calresco.org/sos/sosfaq.htm> seems to me to be rather weak.
>This definition's use of the phrase "higher level" seems problematic,
>too, since it appears that much of the definition's substance rests on
>this phrase, yet that phrase does not seem very meaningful in itself.
True. The FAQ from which it comes is not intended to be definitive
on this point, it is defined in a context of explaining self-organization.
'Higher Level' is defined just prior in the FAQ (equally weakly !) as:
*****
"The system has properties that are emergent, if they are not
intrinsically found within any of the parts, and exist only at a
higher level of description.
2.2 What is a system property ?
When a series of parts are connected into various configurations,
the resultant system no longer solely exhibits the collective properties
of the parts themselves. Instead any additional behaviour attributed
to the system is an example of an emergent system property. A
configuration can be physical, logical or statistical, all can show
unexpected features that cannot be reduced to an additive property
of the individual parts."
*****
Emergence, in the context of this newsgroup anyway, is more often
restricted to apply only to non-artifactual products, i.e. only 'natural'
forms of 'the 'whole is greater than the parts' - a mathematical definition
of this could be 1 + 1 = 3 or some such (i.e. 'rationally' it won't add
up if we use a metaphysical mathematics of disjoint 'objects' only ;-)
If anyone requires better definitions then I'd suggest the relevant
articles by Claus Emmeche and others listed on our Online Papers
page: http://www.calresco.org/papers.htm#sel
Luke mentioned a car as 'not a property', but the property of
being say a 'motorised conveyance' is not problematical to me !
Emergence comes in many forms (as do linguistic definitions ;-),
so even a box qualifies here as it has a property (volume) that is not
exhibited by any of its sides (since abstract planes have area only).
For those of a rather deeper philosophical bent, an interesting PhD
thesis concerning computational emergence and poiesis (becoming),
from both ontological and epistemological perspectives, can be found
at: http://mcs.open.ac.uk/sma78/thesis/thesis.html
'Emergent Behaviour', the original question, is (arguably !) perhaps
that sub-set of emergent properties that exhibit a telelogical (or maybe
teleonomical for neo-Darwinian purists ;-) characteristic (e.g. the
functional movement of an amoeba).
Regards
Chris Lucas
CALResCo Group
http://www.calresco.org/
Moshe Sipper
@Article{ronald99emergence,
author = {E. M. A. Ronald and M. Sipper and M. S. Capcarr\`ere},
title = {Design, Observation, Surprise! {A} Test of Emergence},
journal = {Artificial Life},
year = {1999},
volume = {5},
number = {3},
pages = {225-239},
month = {Summer},
}
-- +----------------------------------------------------------------------+ Dr. Moshe Sipper -- Senior Lecturer Office: Deichmann 125 Department of Computer Science Phones: +972-8-647-7880 (voice) Ben-Gurion University +972-8-647-7650 (fax) P.O. Box 653, Beer-Sheva 84105 E-mail: sip...@cs.bgu.ac.il Israel URL: www.moshesipper.com +----------------------------------------------------------------------+
Josiah Wood
Boids. Boids is a program origionally created by Craig Reynolds that
simulated birds flocking. If seen it mentioned in a few books I've read so
it souldn't be too difficult to find information on.
In fact, MSDN (Microsoft Developer Network) has sample code for Visuall C++
6 called "Boids: DirectX Flocking Sample" that does the same thing. It is
actually a sample to help teach DirectX, so it may be difficult to pull out
the parts that are the flocking code.
Hope it helps.
Philipp A. Djang <pdj...@psl.nmsu.edu> wrote in message
news:9qq61l$hrm$1...@bubba.NMSU.Edu...