[FRIAM] Agar, Abduction

Mon Aug 14 16:37:24 EDT 2006

Michael, 

I think I, too, am a fan of abduction, even though I am not so sure I know what it is.  To me it means the use of metaphors to explain.  A great many years ago, when I was still in the monkey business, I was able to demonstrate that the "social structure" of a monkey "group" was the same, whether one convened it as a whole or only as a series of n(n-1)/2 pairs of monkeys, suggesting that a monkey social group is an aggregate property of the behavior of its pairs.  It was a startling observation, one I did not expect and one I did not altogether trust.  What it suggested is that a group of monkeys, maintained in individual cages, and paired for observation, and who never had physical contact with monkeys outside of those meetings, was a good metaphor (model) for the group operating as a group in the ordinary sense.  

This is an example of a very low level abduction.  Natural selection theory ... the idea that what happens in a breeders barnyard or stable etc. can be taken as a model for what happens in nature ... is an example of a very high level abduction.   Evolution ... the idea that the change in species through time is akin to the ramification of a trees branches at it grows upward to the light .... is another.  Good metaphors stimulate thought and experiment, but a metaphor maker has a deep responsibility to stipulate which parts of his metaphor are facetious ... designed for fun and cognitive promotion, not part of what Mary Brenda Hesse calls "the positive heuristic of the metaphor".  Famous authors of widely read books often get away with ignoring that responsibility, viz, Richard Dawkins and his Selfish Gene.  

So.  Are we talking about the same thing when we talk about abduction?  As a man with a stiff hip, abduction is a concept I can use some help with.  

Nick 

----- Original Message ----- 
From: Michael Agar 
To: nickthompson at earthlink.net
Cc: friam at redfish.com
Sent: 8/14/2006 1:00:53 PM 
Subject: Popper misuse

Hi Nick. I'm actually an abduction fan myself.

I shouldn't have taken Popper's name in vain, since all I really meant was falsification. Validity comes out of the research that precedes a model, the model explores and clarifies a core argument of that research, then in science the argument is put to the test in any number of imaginative ways where procedures are explicit and capable of falsifying or at least complicating the argument. Not everyone's cup of code by a long shot, but one I find useful. I don't think very many people think this way either (:

Mike

On Aug 14, 2006, at 10:36 AM, Nicholas Thompson wrote:

Mike A. writes:

How do we make clear the core of a problem through constructing an 
illustration of our own beliefs and assumptions

and say that's exactly what both great science and great art do. 
Science then has the obligation to challenge it against new instances 
of the problem in the classic Popperian way.

One trouble with Popper is, of course, that people just dont think that
way.  We engage in induction no matter how illogical it may be.  Somebody I
knew had a small animal skin.... ferret or something ... nailed to a board
at one end.  When you petted it, it arched its back, so to speak.  Should
we conclude that that is why cats arch their backs when you pet them???? 
Probably not.  

The other trouble with Popper is, as David Stove pointed out, that EVERY
DEDUCTIVE INFERENCE REQUIRES INDUCTION TO GET IT INTO THE REALM OF
PRACTICAL EXPERIMENT.  So, for instance, as we are busily nailing our live
cat to a board to test our deductive inference, we must assume that all our
operations have the same effects in the live cat and in the ferret skin
case, and this assumption is an INDUCTIVE STEP subject to all of Popper's
doubts about the possibility of induction.  

I think Stove concluded that we just had to suck it up and go back to
making rules for inductive inference, dubious as the whole enterprise is.  

So then the question would be, under what conditions do we accept that when
the simple agents that we send forth to do battle in our models product the
same collective behavior as the apparently real agents we see around us,
that the real agents actually behave by the same underlying rules as the
our created ones?  

Stove wrote a subsequent book on induction, but I havent read it.  Has
anybody???

[Original Message]
From: <friam-request at redfish.com>
To: <friam at redfish.com>
Date: 8/14/2006 12:00:21 PM
Subject: Friam Digest, Vol 38, Issue 29

Send Friam mailing list submissions to
friam at redfish.com

To subscribe or unsubscribe via the World Wide Web, visit
http://redfish.com/mailman/listinfo/friam_redfish.com
or, via email, send a message with subject or body 'help' to
friam-request at redfish.com

You can reach the person managing the list at
friam-owner at redfish.com

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Friam digest..."

Today's Topics:

   1. the odd question (Phil Henshaw) (Nicholas Thompson)
   2. The art of agent-based modeling (Jochen Fromm)
   3. Re: The art of agent-based modeling (Marcus G. Daniels)
   4. Re: The art of agent-based modeling (Jochen Fromm)
   5. Re: The art of agent-based modeling (mgd at santafe.edu)
   6. Re: The art of agent-based modeling (Michael Agar)

----------------------------------------------------------------------

Message: 1
Date: Sun, 13 Aug 2006 23:40:29 -0400
From: "Nicholas Thompson" <nickthompson at earthlink.net>
Subject: [FRIAM] the odd question (Phil Henshaw)
To: "Friam" <Friam at redfish.com>
Message-ID: <380-22006811434029151 at earthlink.net>
Content-Type: text/plain; charset="us-ascii"

Phil, 

I hate it when one of my topics gets dropped, and therefore feel guilty
for being one of the DROPPERS, here. 

Sometimes the discussions get so far reaching  and technical that I am
forced to "pass over them in silence" as Wittegenstein said.  

the only piece of your message that I have anything  nearly competent to
say about is  your  ....

"when modern science took an interest in complex systems it, concentrated
on theory rather than on carefully documenting the physical phenomenon."

I wonder if this isnt a common occcurence in science.  Think of
Evolutionary Biology   Darwinism has a much stronger hand on its theories
than it does on the things those theories explain.  Think for a moment
about  our realtive grasp on "natural selection" and "adaptation".  Natural
selection is supposed to the be "cause" of adaptation, yet we seem to
understand the cause much better than we understand the effect.   Ask an
evolutionary biologist to define adaptation: 90 percent will use the word
natural selection in their definitions, because they dont have clue what
they mean by adaptation.  

Thus, it doesnt surprise me that wise and sophisticated people can talk
about the theory of complexity without having a clue what they mean by it.

I got a group of people to gether at Clark a few years back to start a
research project on emergence in human social groups.  We were NEVER able
to come up with a phenomenon that everybody agreed was an instance of
emergence.     
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
/pipermail/friam_redfish.com/attachments/20060813/be24b22f/attachment-0001.h
tml 

------------------------------

Message: 2
Date: Mon, 14 Aug 2006 14:14:55 +0200
From: "Jochen Fromm" <fromm at vs.uni-kassel.de>
Subject: [FRIAM] The art of agent-based modeling
To: "'The Friday Morning Applied Complexity Coffee Group'"
<friam at redfish.com>
Message-ID: <000701c6bf9b$41129210$976fa8c0 at Toshiba>
Content-Type: text/plain; charset="US-ASCII"

One question I meet again and again if I try to
make meaningful agent-based simulations is:
- How do we simulate the core of a problem
  without merely constructing an illustration
  of our own beliefs and assumptions ?
In other words: How detailed should an agent-based 
simulation be ? If the goal is "to capture the principal 
laws behind the exciting variety of new phenomena that become 
apparent when the many units of a complex system interact", as 
Tamas Vicsek says in http://angel.eltehu/~vicsek/images/complex.pdf
then how do we design models that are complex enough but not too 
complex ?

-If the simulation is too simple and matches your 
 own theoretical ideas, then no matter how good these 
 ideas are it is always easy to criticize that the 
 simulation is either not realistic enough or only 
 constructed to illustrate your own ideas and assumptions. 
-If the simulation is too complex and matches 
 official experimental data, everything takes a 
 lot amount of time (creation, setup and execution of
 the experiment and finally the cumbersome analysis
 of the complex outcomes), and it becomes increasingly 
 difficult to identify the principal laws, because it is 
 easy to get lost in the data or bogged down in details

The "art of agent-based modeling" looks really like an art 
to me, something only mastered by a few scientists (for instance 
Axelrod). Grimm et al. propose 'pattern-oriented modeling',
Macy and Willer say "Keep it simple" and "Test validity".
What do you think is the best solution for this problem ?

Macy and Willer
"From Factors to Actors: Computational Sociology and Agent-Based Modeling"

http://www.casos.cs.cmu.edu/education/phd/classpapers/Macy_Factors_2001.pdf

Grimm et al. 
"Pattern-oriented modeling of agent-based complex systems"
Science Vol. 310. no. 5750 (2005) 987-991
http://www.ufz.de/index.php?de=4976

-J.

-----Original Message-----
From: Michael Agar
Sent: Saturday, August 12, 2006 5:05 PM
To: The Friday Morning Applied Complexity Coffee Group
Subject: [FRIAM] complexity and society

[...] If you are considering a model, I like Axelrod's way of thinking  
about them. He sees them as "thought experiment labs" for a  
conclusion based on social research. So first of all the social  
research has to be solid to really do it properly. More often than  
not it isn't.

The lab let's you test arguments of the form, if people do things in  
particular ways properties will emerge at the level of society. By  
"test" I mean it lets you see if the conclusion can be "generated,"  
to use Epstein and Axtell's concept, in just the way your social  
research suggests that it can. It's a way of making the argument that  
underlies the conclusion explicit so it can be better evaluated, and  
it allows for exploration of the space of results that the same  
argument produces and alternative spaces given control parameter  
changes. It's a test of plausibility and an exercise in clarity,  
nothing more, nothing less. [...]

------------------------------

Message: 3
Date: Mon, 14 Aug 2006 08:07:05 -0600
From: "Marcus G. Daniels" <mgd at santafe.edu>
Subject: Re: [FRIAM] The art of agent-based modeling
To: The Friday Morning Applied Complexity Coffee Group
<friam at redfish.com>
Message-ID: <44E08389.5070109 at santafe.edu>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Jochen,
-If the simulation is too complex and matches 
 official experimental data, everything takes a 
 lot amount of time (creation, setup and execution of
 the experiment and finally the cumbersome analysis
 of the complex outcomes), and it becomes increasingly 
 difficult to identify the principal laws, because it is 
 easy to get lost in the data or bogged down in details

This may be a false choice.   In the case of having some data of 
moderate resolution, there's no point in making a hugely elaborate model 
and simulation, because you'll never be able to validate beyond your 
data anyway.   And if you don't validate, although the modeling still 
may be useful as an thought experiment, it isn't science.  You have to 
be able to say something that can be shown to be wrong.   If you do aim 
to learn things about the world and then predict them it's not desirable 
to have giant black box with lots of moving parts.   It's better, if at 
all possible, to have a simple story and make the simulation nothing 
more than apparatus to help extend the data so that the dynamics can be 
studied by theoreticians.

Another mode of use for ABMs is to lower expectations of theoretical 
traction and opportunistically look for ways a model makes useful 
predictions and then modify the model in that direction over time.   
This is a risky and expensive craft, but one that might have high enough 
payoffs to consider (e.g. national security).

It depends on the data and what is of interest.   If the data tells you 
about a number of rare events, and it is these events is what you really 
care about, then it may make sense to loosely model everyday behaviors 
and focus on model microstructure that can create the rare events you 
care about.

Finally, sometimes microstructure is known with clearly defined degrees 
of freedom, and the dynamics are of interest.  Consider modeling a 
factory where different assembly regimes are to be evaluated..  There's 
no need to validate here because the whole exercise is to answer 
what-ifs about realizable specific systems.

Marcus

------------------------------

Message: 4
Date: Mon, 14 Aug 2006 17:04:40 +0200
From: "Jochen Fromm" <fromm at vs.uni-kassel.de>
Subject: Re: [FRIAM] The art of agent-based modeling
To: "'The Friday Morning Applied Complexity Coffee Group'"
<friam at redfish.com>
Message-ID: <000801c6bfb2$f76ede80$976fa8c0 at Toshiba>
Content-Type: text/plain; charset="US-ASCII"

Of course it is the essence of science to verify hypotheses
by experiments. Yet sometimes we have neither suitable 
experimental data nor a solid theory, for example
in the case of very large agent-based systems (for instance 
for the self-organization and self-management of large 
internet applications on planetary scale, or the modeling 
of historical processes with millions of actors). It is 
hardly possible to examine these systems without simplified
models, and in this case the questions I mentioned seem to
be justified.

In traditional "factor-based" or "equation-based modeling" 
we use differential equations and everything is based
on a soild theory: mathematics. This traditional modeling 
has a century-long history and we know the suitable parameters,
equations and models. Agent-based modeling has a short history, 
we don't know exactly the suitable parameters, agents and models, 
and worst of all it is not based on a solid theoretical 
theorem-lemma-proof science or calculus like mathematics.

What is missing is a solid science of ABM or a new science of
complexity - something in the direction of Wolfram's NKS idea
(exploring computational universes in a systematic way). Just
as formal, symmetrical and regular systems can be described by 
mathematics and 'equation-based modeling', complex systems can 
in principle be described by a 'NKS' and agent-based modeling 
- which seems to be more an art than a science. 

-J.

------------------------------

Message: 5
Date: Mon, 14 Aug 2006 09:50:59 -0600
From: mgd at santafe.edu
Subject: Re: [FRIAM] The art of agent-based modeling
To: The Friday Morning Applied Complexity Coffee Group
<friam at redfish.com>
Message-ID: <1155570659.44e09be35f52a at webmail.santafe.edu>
Content-Type: text/plain; charset=ISO-8859-1

Quoting Jochen Fromm <fromm at vs.uni-kassel.de>:

Just as formal, symmetrical and regular systems can be described by 
mathematics and 'equation-based modeling', complex systems can 
in principle be described by a 'NKS' and agent-based modeling 
- which seems to be more an art than a science. 

In context, I think is a verification issue.  

ABMs are useful for poking around a complicated system to see what
matters and 
what doesn't by using a familiar and direct way of describing things, and
to 
leave the abstractions for later.  ABMs complement traditional techniques
of 
analysis by extending data.

The imperative programming languages that are typically used to make the 
simulations are prone to a variety of programming mistakes but the
continue to 
be used because 1) they are common and 2) they provide an easy way to
think 
about side effects (e.g. modifications to a landscape).

Equation-based modelling is more like functional programming, e.g.
programming 
languages like Haskell that are side-effect free.   I see ABMs moving to
these 
kinds of programming languages so that components of a simulation can be
shown 
to be correct, and preferably by automated means.  

As a practical matter, I think it isn't a big deal.  Unit testing during 
development by experienced programmers/modelers does a good job of
shaking out 
bugs.

Marcus

------------------------------

Message: 6
Date: Mon, 14 Aug 2006 09:58:05 -0600
From: Michael Agar <magar at anth.umd.edu>
Subject: Re: [FRIAM] The art of agent-based modeling
To: The Friday Morning Applied Complexity Coffee Group
<friam at redfish.com>
Message-ID: <18528635-4305-4EB8-8D01-CA9C885E3826 at anth.umd.edu>
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed

On Aug 14, 2006, at 6:14 AM, Jochen Fromm wrote:

- How do we simulate the core of a problem
  without merely constructing an illustration
  of our own beliefs and assumptions ?

I'd change this to

How do we make clear the core of a problem through constructing an  
illustration of our own beliefs and assumptions

and say that's exactly what both great science and great art do.  
Science then has the obligation to challenge it against new instances  
of the problem in the classic Popperian way.

Mike

------------------------------

_______________________________________________
Friam mailing list
Friam at redfish.com
http://redfish.com/mailman/listinfo/friam_redfish.com

End of Friam Digest, Vol 38, Issue 29
*************************************
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/friam_redfish.com/attachments/20060814/b0d72025/attachment-0001.html 