Friam

Re: Agents, stocks, and flows

Classic

List

Threaded

20 messages Options

Russ Abbott

Re: Agents, stocks, and flows

In a discussion with a colleague today we talked briefly about stocks and flows networks. It struck me that a stocks and flows model is a limited sort of service-oriented agent-based model. In a service-oriented agent-based model, agents accept inputs and produce outputs -- the simplest version being a supply chain. That's really a stocks and flows model in which the agents control the flows. Important differences are:

In an agent-based model, the agents are assumed to be autonomous in various ways. In a stocks and flows model the flow rates are not autonomous. The flow rates are equations that don't have the ability to change themselves.They are assumed to be facts about the nature of the domain being modeled.

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

So this raises the question whether anyone knows of any work in stocks and flows modeling that addresses stocks and flows networks that are flexible in the ways just mentioned.

-- Russ

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Nick Thompson

Re: Agents, stocks, and flows

All,

I have tried to stay out of this discussion because I know even less about this subject than usual, if such a thing is possible. But .....

Russ Abbott wrote:

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

Which led me to wonder if there is any thing lurking in the notion of "self-reconfiguration" here that might be making the programing more difficult. What would be lost (or gained) if we replaced the words "ability to reconfigure themselves" with the words "sometimes are reconfigured". That's the thing about metaphysics: cant live with it; cant live without it. At the risk of saying something both controversial and incomprehensible, isnt the notion of self-organization literally a non starter because, in control theory, the control parameter, the measure by which the system takes stock of its own organization, is always some part or feature of the system, not the whole system. It's a cue.

all be best,

Nick

Nicholas S. Thompson

Emeritus Professor of Psychology and Ethology,

Clark University ([hidden email])

http://home.earthlink.net/~nickthompson/naturaldesigns/

----- Original Message -----

From: [hidden email]

To: [hidden email]
Cc: [hidden email]; [hidden email]; [hidden email]; [hidden email]

Sent: 8/28/2009 5:49:40 AM

Subject: Re: [FRIAM] Agents, stocks, and flows

In a discussion with a colleague today we talked briefly about stocks and flows networks. It struck me that a stocks and flows model is a limited sort of service-oriented agent-based model. In a service-oriented agent-based model, agents accept inputs and produce outputs -- the simplest version being a supply chain. That's really a stocks and flows model in which the agents control the flows. Important differences are:

In an agent-based model, the agents are assumed to be autonomous in various ways. In a stocks and flows model the flow rates are not autonomous. The flow rates are equations that don't have the ability to change themselves.They are assumed to be facts about the nature of the domain being modeled.

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.
So this raises the question whether anyone knows of any work in stocks and flows modeling that addresses stocks and flows networks that are flexible in the ways just mentioned.

-- Russ

Russ Abbott

Re: Agents, stocks, and flows

My question was simpler than that. Is there work on stocks and flows models in which the network structure is not static?

-- Russ

On Fri, Aug 28, 2009 at 6:59 AM, Nicholas Thompson <[hidden email]> wrote:

All,

I have tried to stay out of this discussion because I know even less about this subject than usual, if such a thing is possible. But .....

Russ Abbott wrote:

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

Which led me to wonder if there is any thing lurking in the notion of "self-reconfiguration" here that might be making the programing more difficult. What would be lost (or gained) if we replaced the words "ability to reconfigure themselves" with the words "sometimes are reconfigured". That's the thing about metaphysics: cant live with it; cant live without it. At the risk of saying something both controversial and incomprehensible, isnt the notion of self-organization literally a non starter because, in control theory, the control parameter, the measure by which the system takes stock of its own organization, is always some part or feature of the system, not the whole system. It's a cue.

all be best,

Nick

Nicholas S. Thompson

Emeritus Professor of Psychology and Ethology,

Clark University ([hidden email])

http://home.earthlink.net/~nickthompson/naturaldesigns/

----- Original Message -----

From: [hidden email]

To: [hidden email]
Cc: [hidden email]; [hidden email]; [hidden email]; [hidden email]

Sent: 8/28/2009 5:49:40 AM

Subject: Re: [FRIAM] Agents, stocks, and flows

In a discussion with a colleague today we talked briefly about stocks and flows networks. It struck me that a stocks and flows model is a limited sort of service-oriented agent-based model. In a service-oriented agent-based model, agents accept inputs and produce outputs -- the simplest version being a supply chain. That's really a stocks and flows model in which the agents control the flows. Important differences are:

In an agent-based model, the agents are assumed to be autonomous in various ways. In a stocks and flows model the flow rates are not autonomous. The flow rates are equations that don't have the ability to change themselves.They are assumed to be facts about the nature of the domain being modeled.

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

So this raises the question whether anyone knows of any work in stocks and flows modeling that addresses stocks and flows networks that are flexible in the ways just mentioned.

-- Russ

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Douglas Roberts-2

Re: Agents, stocks, and flows

Why are you asking this? Do you want to build an ABM of a system that models stock transactions, or write a paper about how do do so, of just find a place for this class of simulation that fits your worldview?

--Doug

On Fri, Aug 28, 2009 at 8:26 AM, Russ Abbott <[hidden email]> wrote:

My question was simpler than that. Is there work on stocks and flows models in which the network structure is not static?

-- Russ

On Fri, Aug 28, 2009 at 6:59 AM, Nicholas Thompson <[hidden email]> wrote:

All,

I have tried to stay out of this discussion because I know even less about this subject than usual, if such a thing is possible. But .....

Russ Abbott wrote:

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

Which led me to wonder if there is any thing lurking in the notion of "self-reconfiguration" here that might be making the programing more difficult. What would be lost (or gained) if we replaced the words "ability to reconfigure themselves" with the words "sometimes are reconfigured". That's the thing about metaphysics: cant live with it; cant live without it. At the risk of saying something both controversial and incomprehensible, isnt the notion of self-organization literally a non starter because, in control theory, the control parameter, the measure by which the system takes stock of its own organization, is always some part or feature of the system, not the whole system. It's a cue.

all be best,

Nick

Nicholas S. Thompson

Emeritus Professor of Psychology and Ethology,

Clark University ([hidden email])

http://home.earthlink.net/~nickthompson/naturaldesigns/

----- Original Message -----

From: [hidden email]

To: [hidden email]
Cc: [hidden email]; [hidden email]; [hidden email]; [hidden email]

Sent: 8/28/2009 5:49:40 AM

Subject: Re: [FRIAM] Agents, stocks, and flows

In a discussion with a colleague today we talked briefly about stocks and flows networks. It struck me that a stocks and flows model is a limited sort of service-oriented agent-based model. In a service-oriented agent-based model, agents accept inputs and produce outputs -- the simplest version being a supply chain. That's really a stocks and flows model in which the agents control the flows. Important differences are:

In an agent-based model, the agents are assumed to be autonomous in various ways. In a stocks and flows model the flow rates are not autonomous. The flow rates are equations that don't have the ability to change themselves.They are assumed to be facts about the nature of the domain being modeled.

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

So this raises the question whether anyone knows of any work in stocks and flows modeling that addresses stocks and flows networks that are flexible in the ways just mentioned.

-- Russ

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Russ Abbott

Re: Agents, stocks, and flows

The other day I described a framework (service-oriented agent-based modeling) that I thought would suit the kinds of M&S I want to do. As far as I know there isn't any literature on this sort of modeling paradigm. But if a re-interpretation of stocks and flows modeling can be understood in that light and if work has already been done in that area, I want to know about it so that I don't reproduce that effort.

-- Russ

On Fri, Aug 28, 2009 at 7:42 AM, Douglas Roberts <[hidden email]> wrote:

Why are you asking this? Do you want to build an ABM of a system that models stock transactions, or write a paper about how do do so, of just find a place for this class of simulation that fits your worldview?

--Doug

On Fri, Aug 28, 2009 at 8:26 AM, Russ Abbott <[hidden email]> wrote:
My question was simpler than that. Is there work on stocks and flows models in which the network structure is not static?

-- Russ

On Fri, Aug 28, 2009 at 6:59 AM, Nicholas Thompson <[hidden email]> wrote:

All,

I have tried to stay out of this discussion because I know even less about this subject than usual, if such a thing is possible. But .....

Russ Abbott wrote:

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

Which led me to wonder if there is any thing lurking in the notion of "self-reconfiguration" here that might be making the programing more difficult. What would be lost (or gained) if we replaced the words "ability to reconfigure themselves" with the words "sometimes are reconfigured". That's the thing about metaphysics: cant live with it; cant live without it. At the risk of saying something both controversial and incomprehensible, isnt the notion of self-organization literally a non starter because, in control theory, the control parameter, the measure by which the system takes stock of its own organization, is always some part or feature of the system, not the whole system. It's a cue.

all be best,

Nick

Nicholas S. Thompson

Emeritus Professor of Psychology and Ethology,

Clark University ([hidden email])

http://home.earthlink.net/~nickthompson/naturaldesigns/

----- Original Message -----

From: [hidden email]

To: [hidden email]
Cc: [hidden email]; [hidden email]; [hidden email]; [hidden email]

Sent: 8/28/2009 5:49:40 AM

Subject: Re: [FRIAM] Agents, stocks, and flows

In a discussion with a colleague today we talked briefly about stocks and flows networks. It struck me that a stocks and flows model is a limited sort of service-oriented agent-based model. In a service-oriented agent-based model, agents accept inputs and produce outputs -- the simplest version being a supply chain. That's really a stocks and flows model in which the agents control the flows. Important differences are:

In an agent-based model, the agents are assumed to be autonomous in various ways. In a stocks and flows model the flow rates are not autonomous. The flow rates are equations that don't have the ability to change themselves.They are assumed to be facts about the nature of the domain being modeled.

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

So this raises the question whether anyone knows of any work in stocks and flows modeling that addresses stocks and flows networks that are flexible in the ways just mentioned.

-- Russ

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Stephen Guerin

Re: Agents, stocks, and flows

Netlogo includes a stocks and flows component (system dynamics
modeler) <http://ccl.northwestern.edu/netlogo/docs/systemdynamics.html/
>. I don't believe the network of stocks and flows is reconfigurable
at runtime though it does have a basic integration with the agent-
based modeling side of Netlogo which allows for some cool
possibilities. So a stock can represent some aggregated property of a
dynamically changing agent population. And going the other direction,
the agent population can react to some changing flow or stock at the
SD level.

In addition to the dynamic nature of the interaction networks, I think
SD models differ from ABM mostly as they tend to be models of more
aggregate properties of a system whereas ABM are more micro looking at
agent interactions that give rise to macro stocks and flows. There are
nice "docking" experiments out there that look at the common results
and the advantage of each approach modeling the same system. Lotka-
Volterra (wolf/sheep), Infection models (SIR), and supply chain are
three of the most common model types I've seen.

-S

--- -. . ..-. .. ... .... - .-- --- ..-. .. ... ....
[hidden email]
(m) 505.577.5828 (o) 505.995.0206
redfish.com _ sfcomplex.org _ simtable.com _ lava3d.com

On Aug 28, 2009, at 9:02 AM, Russ Abbott wrote:

> The other day I described a framework (service-oriented agent-based
> modeling) that I thought would suit the kinds of M&S I want to do.
> As far as I know there isn't any literature on this sort of modeling
> paradigm. But if a re-interpretation of stocks and flows modeling
> can be understood in that light and if work has already been done in
> that area, I want to know about it so that I don't reproduce that
> effort.
>
> -- Russ
>
>
>
> On Fri, Aug 28, 2009 at 7:42 AM, Douglas Roberts <doug@parrot-
> farm.net> wrote:
> Why are you asking this? Do you want to build an ABM of a system
> that models stock transactions, or write a paper about how do do so,
> of just find a place for this class of simulation that fits your
> worldview?
>
> --Doug
>
>
> On Fri, Aug 28, 2009 at 8:26 AM, Russ Abbott <[hidden email]>
> wrote:
> My question was simpler than that. Is there work on stocks and flows
> models in which the network structure is not static?
>
> -- Russ
>
>
>
> On Fri, Aug 28, 2009 at 6:59 AM, Nicholas Thompson <[hidden email]
> > wrote:
> All,
>
> I have tried to stay out of this discussion because I know even less
> about this subject than usual, if such a thing is possible. But .....
>
> Russ Abbott wrote:
>
> In a service-oriented agent-based model the agents have the ability
> to reconfigure themselves dynamically and perhaps even to add new
> agents and new stock nodes. In a stocks and flows model, the
> structure of the network static.
>
> Which led me to wonder if there is any thing lurking in the notion
> of "self-reconfiguration" here that might be making the programing
> more difficult. What would be lost (or gained) if we replaced the
> words "ability to reconfigure themselves" with the words "sometimes
> are reconfigured". That's the thing about metaphysics: cant live
> with it; cant live without it. At the risk of saying something both
> controversial and incomprehensible, isnt the notion of self-
> organization literally a non starter because, in control theory, the
> control parameter, the measure by which the system takes stock of
> its own organization, is always some part or feature of the system,
> not the whole system. It's a cue.
>
> all be best,
>
> Nick
>
> Nicholas S. Thompson
> Emeritus Professor of Psychology and Ethology,
> Clark University ([hidden email])
> http://home.earthlink.net/~nickthompson/naturaldesigns/
>
>
>
>
> ----- Original Message -----
> From: Russ Abbott
> To: The Friday Morning Applied Complexity Coffee Group
> Cc: Antony W. Iorio; [hidden email]; [hidden email]
> ; Lowe,Donald
> Sent: 8/28/2009 5:49:40 AM
> Subject: Re: [FRIAM] Agents, stocks, and flows
>
> In a discussion with a colleague today we talked briefly about
> stocks and flows networks. It struck me that a stocks and flows
> model is a limited sort of service-oriented agent-based model. In a
> service-oriented agent-based model, agents accept inputs and produce
> outputs -- the simplest version being a supply chain. That's really
> a stocks and flows model in which the agents control the flows.
> Important differences are:
> • In an agent-based model, the agents are assumed to be autonomous
> in various ways. In a stocks and flows model the flow rates are not
> autonomous. The flow rates are equations that don't have the ability
> to change themselves.They are assumed to be facts about the nature
> of the domain being modeled.
> • In a service-oriented agent-based model the agents have the
> ability to reconfigure themselves dynamically and perhaps even to
> add new agents and new stock nodes. In a stocks and flows model, the
> structure of the network static.
> So this raises the question whether anyone knows of any work in
> stocks and flows modeling that addresses stocks and flows networks
> that are flexible in the ways just mentioned.
>
> -- Russ
>
>
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> lectures, archives, unsubscribe, maps at http://www.friam.org
>
>
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> lectures, archives, unsubscribe, maps at http://www.friam.org
>
>
>
>
>
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> lectures, archives, unsubscribe, maps at http://www.friam.org

glen e. p. ropella-2

Re: Agents, stocks, and flows

In reply to this post by Russ Abbott

Thus spake Russ Abbott circa 09-08-28 07:26 AM:
> My question was simpler than that. Is there work on stocks and flows models
> in which the network structure is not static?

I suspect you won't find a toolkit that does this because the diagrams
are (usually?) animated via a system of differential equations, which
are difficult to change while being integrated. If the underlying
implementation were discretized at the level of the stock and flow unit,
then it would be relatively trivial. But, then, what's the point of a
stock and flow model if, underneath, it's not a diff eq system?

--
glen e. p. ropella, 971-222-9095, http://agent-based-modeling.com

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Roger Critchlow-2

Re: Agents, stocks, and flows

In reply to this post by Russ Abbott

I don't think you'll find this because it implies programming a higher purpose and allowing the agents to jump the rails, as it were, and start negotiating their way through the combinatorics of alternative networks. Similarly, you won't find models in which agents invent new inputs to monitor, new outputs to generate, and new rules which involve new inputs and new outputs.

Optimization within a fixed solution space, which is what we do when we let agents play with the flow through a fixed network or let them search out the most profitable rules in a set of prespecified alternatives, gets hairy enough without opening things up to the infinity of potential solutions that we didn't have time to program into the model ourselves.

Neal Stephenson's novel Anathem has some discussions, central to the plot, about how the human mind filters the combinatorically possible down to the mechanically feasible and further down to the set of outcomes worth thinking about.

It's not obvious how to program the same capabilities without the teaching the agents how to apply the same common sense and expert senses which the programmer uses to frame a fixed solution space.

-- rec --

On Fri, Aug 28, 2009 at 6:48 AM, Russ Abbott <[hidden email]> wrote:

In a discussion with a colleague today we talked briefly about stocks and flows networks. It struck me that a stocks and flows model is a limited sort of service-oriented agent-based model. In a service-oriented agent-based model, agents accept inputs and produce outputs -- the simplest version being a supply chain. That's really a stocks and flows model in which the agents control the flows. Important differences are:

In an agent-based model, the agents are assumed to be autonomous in various ways. In a stocks and flows model the flow rates are not autonomous. The flow rates are equations that don't have the ability to change themselves.They are assumed to be facts about the nature of the domain being modeled.

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

So this raises the question whether anyone knows of any work in stocks and flows modeling that addresses stocks and flows networks that are flexible in the ways just mentioned.

-- Russ

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Marcus G. Daniels

Re: Agents, stocks, and flows

Roger Critchlow wrote:
> It's not obvious how to program the same capabilities without the
> teaching the agents how to apply the same common sense and expert
> senses which the programmer uses to frame a fixed solution space.
Framing a fixed solution space is tricky! Take a data set with lots of
possible signals and chop it in two pieces. If one set gives you a
variable with states A, B and the other set has the same variables with
only C and D then a model that predicts on distinctions between A and B
won't be useful. Using the other half of the data to inform what can
be considered a stable signal is in some sense cheating as much as
fitting a model to all of the data and then testing using it too.

Marcus

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Russell Standish

Re: Agents, stocks, and flows

In reply to this post by Roger Critchlow-2

On Fri, Aug 28, 2009 at 12:46:26PM -0500, Roger Critchlow wrote:

> I don't think you'll find this because it implies programming a higher
> purpose and allowing the agents to jump the rails, as it were, and start
> negotiating their way through the combinatorics of alternative networks.
> Similarly, you won't find models in which agents invent new inputs to
> monitor, new outputs to generate, and new rules which involve new inputs and
> new outputs.
>
> Optimization within a fixed solution space, which is what we do when we let
> agents play with the flow through a fixed network or let them search out the
> most profitable rules in a set of prespecified alternatives, gets hairy
> enough without opening things up to the infinity of potential solutions that
> we didn't have time to program into the model ourselves.
>

EcoLab is an example of a model where the state space evolves over
time rather than staying fixed. It is not quite the ABM that Russ
Abbott is looking for, but does illustrate that it is possible. One
crucial feature is that there must be a separation of scales - the
dynamics of the system (optimisation, or whatever) must occur more
rapidly than the change to the state space. Otherwise, you get what is
known in the ALife world as a mutational meltdown - evolution ceases
to operate.

--

----------------------------------------------------------------------------
Prof Russell Standish Phone 0425 253119 (mobile)
Mathematics
UNSW SYDNEY 2052 [hidden email]
Australia http://www.hpcoders.com.au
----------------------------------------------------------------------------

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Douglas Roberts-2

Re: Agents, stocks, and flows

Has anybody just tried to design this application the old-fashioned way; i.e., develop a set of requirements that

define the interactions between the components of the system,
identify (clearly, no vagueness allowed) the desired results from running the simulation,
identify (clearly, no vagueness allowed) the inputs for the simulation, and

*then* determine what design best fits the application?

Just asking, 'cause this thread so far sniffs out suspiciously like another "I want to talk about how *I* want to think about how (in the purest theoretical sense) simulations should be designed/implemented/thought of."

Just asking...

--Doug

As compared to endlessly seeing that special, elegant simulation implementation system that is the ideal match to this particular problem domain?

On Fri, Aug 28, 2009 at 6:26 PM, russell standish <[hidden email]> wrote:

On Fri, Aug 28, 2009 at 12:46:26PM -0500, Roger Critchlow wrote:
> I don't think you'll find this because it implies programming a higher
> purpose and allowing the agents to jump the rails, as it were, and start
> negotiating their way through the combinatorics of alternative networks.
> Similarly, you won't find models in which agents invent new inputs to
> monitor, new outputs to generate, and new rules which involve new inputs and
> new outputs.
>
> Optimization within a fixed solution space, which is what we do when we let
> agents play with the flow through a fixed network or let them search out the
> most profitable rules in a set of prespecified alternatives, gets hairy
> enough without opening things up to the infinity of potential solutions that
> we didn't have time to program into the model ourselves.
>

EcoLab is an example of a model where the state space evolves over
time rather than staying fixed. It is not quite the ABM that Russ
Abbott is looking for, but does illustrate that it is possible. One
crucial feature is that there must be a separation of scales - the
dynamics of the system (optimisation, or whatever) must occur more
rapidly than the change to the state space. Otherwise, you get what is
known in the ALife world as a mutational meltdown - evolution ceases
to operate.

--

----------------------------------------------------------------------------
Prof Russell Standish Phone 0425 253119 (mobile)
Mathematics
UNSW SYDNEY 2052 [hidden email]
Australia http://www.hpcoders.com.au
----------------------------------------------------------------------------

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Steve Smith

Re: Agents, stocks, and flows

In reply to this post by Roger Critchlow-2

Roger -

I think this is a good analysis of Russ's idea/question/pursuit.   But/And that is what makes it an interesting pursuit.   Self-modifying code has always been a hairy-scary mess but/and we have slowly found ways to tame it enough to make it useful in various situations.

Two platitudes I've heard coming out of complexity science :
   1) Constraint provides form
   2) When stuck solving a problem, add another level of indirection

The first platitude refers to why there might not be many/any ABM systems which offer self-modification (as Roger alludes to here)... so far, nobody has been able to release the constraint and still get enough coherence out to be useful/interesting.

The second platitude refers to why one might *want to* do what Russ wants to. It would seem that this form of self-modification is a new level of indirection.

I wonder if the combined genius, memory, attention, focus of this group can come up with real-world systems which (obviously) seem to work this way (require this level of abstraction/complexity to model).

-Steve

I don't think you'll find this because it implies programming a higher purpose and allowing the agents to jump the rails, as it were, and start negotiating their way through the combinatorics of alternative networks. Similarly, you won't find models in which agents invent new inputs to monitor, new outputs to generate, and new rules which involve new inputs and new outputs.

Optimization within a fixed solution space, which is what we do when we let agents play with the flow through a fixed network or let them search out the most profitable rules in a set of prespecified alternatives, gets hairy enough without opening things up to the infinity of potential solutions that we didn't have time to program into the model ourselves.

Neal Stephenson's novel Anathem has some discussions, central to the plot, about how the human mind filters the combinatorically possible down to the mechanically feasible and further down to the set of outcomes worth thinking about.

It's not obvious how to program the same capabilities without the teaching the agents how to apply the same common sense and expert senses which the programmer uses to frame a fixed solution space.

-- rec --

On Fri, Aug 28, 2009 at 6:48 AM, Russ Abbott <[hidden email]> wrote:

In a discussion with a colleague today we talked briefly about stocks and flows networks. It struck me that a stocks and flows model is a limited sort of service-oriented agent-based model. In a service-oriented agent-based model, agents accept inputs and produce outputs -- the simplest version being a supply chain. That's really a stocks and flows model in which the agents control the flows. Important differences are:

In an agent-based model, the agents are assumed to be autonomous in various ways. In a stocks and flows model the flow rates are not autonomous. The flow rates are equations that don't have the ability to change themselves.They are assumed to be facts about the nature of the domain being modeled.

In a service-oriented agent-based model the agents have the ability to reconfigure themselves dynamically and perhaps even to add new agents and new stock nodes. In a stocks and flows model, the structure of the network static.

So this raises the question whether anyone knows of any work in stocks and flows modeling that addresses stocks and flows networks that are flexible in the ways just mentioned.

-- Russ

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Douglas Roberts-2

Re: Agents, stocks, and flows

On Sat, Aug 29, 2009 at 10:53 AM, Steve Smith <[hidden email]> wrote:

I wonder if the combined genius, memory, attention, focus of this group can come up with real-world systems which (obviously) seem to work this way (require this level of abstraction/complexity to model).

-Steve

I thought I suggested this yesterday...

"Has anybody just tried to design this application the old-fashioned way; i.e., develop a set of requirements that

define the interactions between the components of the system,
identify (clearly, no vagueness allowed) the desired results from running the simulation,
identify (clearly, no vagueness allowed) the inputs for the simulation, and

*then* determine what design best fits the application?

[Blah Blah Blah]"

--Doug

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Steve Smith

Re: Agents, stocks, and flows

In reply to this post by Douglas Roberts-2

Doug -

Suzanne and I just watched Paper Chase (1973) again and were treated to the dry commentary of John Hausman as curmudgeonly professor Kingsfield.    Your comment reminded me of the equally dry delivery he gave in a TV commercial several years later for an investment house? where he uttered the line

    "We make money the old-fashioned way, we EARN it!"

That noted, I would submit that the FRIAM list (and Friday kaffe klatch) *IS* more about speculative discussion than it is about problem solving. I find there is a delicate distinction between the interesting exploration of ideas through group speculation and well... what I think we all know of as a "circle jerk". I take this risk every time I let myself get drawn into a thread.

- Steve

Has anybody just tried to design this application the old-fashioned way; i.e., develop a set of requirements that

define the interactions between the components of the system,

identify (clearly, no vagueness allowed) the desired results from running the simulation,

identify (clearly, no vagueness allowed) the inputs for the simulation, and

*then* determine what design best fits the application?

Just asking, 'cause this thread so far sniffs out suspiciously like another "I want to talk about how *I* want to think about how (in the purest theoretical sense) simulations should be designed/implemented/thought of."

Just asking...

--Doug

As compared to endlessly seeing that special, elegant simulation implementation system that is the ideal match to this particular problem domain?

On Fri, Aug 28, 2009 at 6:26 PM, russell standish <[hidden email]> wrote:

On Fri, Aug 28, 2009 at 12:46:26PM -0500, Roger Critchlow wrote:
> I don't think you'll find this because it implies programming a higher
> purpose and allowing the agents to jump the rails, as it were, and start
> negotiating their way through the combinatorics of alternative networks.
> Similarly, you won't find models in which agents invent new inputs to
> monitor, new outputs to generate, and new rules which involve new inputs and
> new outputs.
>
> Optimization within a fixed solution space, which is what we do when we let
> agents play with the flow through a fixed network or let them search out the
> most profitable rules in a set of prespecified alternatives, gets hairy
> enough without opening things up to the infinity of potential solutions that
> we didn't have time to program into the model ourselves.
>

EcoLab is an example of a model where the state space evolves over
time rather than staying fixed. It is not quite the ABM that Russ
Abbott is looking for, but does illustrate that it is possible. One
crucial feature is that there must be a separation of scales - the
dynamics of the system (optimisation, or whatever) must occur more
rapidly than the change to the state space. Otherwise, you get what is
known in the ALife world as a mutational meltdown - evolution ceases
to operate.

--

----------------------------------------------------------------------------
Prof Russell Standish Phone 0425 253119 (mobile)
Mathematics
UNSW SYDNEY 2052 [hidden email]
Australia http://www.hpcoders.com.au
----------------------------------------------------------------------------

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Douglas Roberts-2

Re: Agents, stocks, and flows

Steve,

I confess that I enjoy rubbing their noses in their speculative circle-jerks, now and then.

Another topic: do you know Ed MacKerrow?

--Doug

On Sat, Aug 29, 2009 at 11:06 AM, Steve Smith <[hidden email]> wrote:

Doug -

Suzanne and I just watched Paper Chase (1973) again and were treated to the dry commentary of John Hausman as curmudgeonly professor Kingsfield.    Your comment reminded me of the equally dry delivery he gave in a TV commercial several years later for an investment house? where he uttered the line

    "We make money the old-fashioned way, we EARN it!"

That noted, I would submit that the FRIAM list (and Friday kaffe klatch) *IS* more about speculative discussion than it is about problem solving. I find there is a delicate distinction between the interesting exploration of ideas through group speculation and well... what I think we all know of as a "circle jerk". I take this risk every time I let myself get drawn into a thread.

- Steve
Has anybody just tried to design this application the old-fashioned way; i.e., develop a set of requirements that

define the interactions between the components of the system,

identify (clearly, no vagueness allowed) the desired results from running the simulation,

identify (clearly, no vagueness allowed) the inputs for the simulation, and

*then* determine what design best fits the application?

Just asking, 'cause this thread so far sniffs out suspiciously like another "I want to talk about how *I* want to think about how (in the purest theoretical sense) simulations should be designed/implemented/thought of."

Just asking...

--Doug

As compared to endlessly seeing that special, elegant simulation implementation system that is the ideal match to this particular problem domain?

On Fri, Aug 28, 2009 at 6:26 PM, russell standish <[hidden email]> wrote:

On Fri, Aug 28, 2009 at 12:46:26PM -0500, Roger Critchlow wrote:
> I don't think you'll find this because it implies programming a higher
> purpose and allowing the agents to jump the rails, as it were, and start
> negotiating their way through the combinatorics of alternative networks.
> Similarly, you won't find models in which agents invent new inputs to
> monitor, new outputs to generate, and new rules which involve new inputs and
> new outputs.
>
> Optimization within a fixed solution space, which is what we do when we let
> agents play with the flow through a fixed network or let them search out the
> most profitable rules in a set of prespecified alternatives, gets hairy
> enough without opening things up to the infinity of potential solutions that
> we didn't have time to program into the model ourselves.
>

EcoLab is an example of a model where the state space evolves over
time rather than staying fixed. It is not quite the ABM that Russ
Abbott is looking for, but does illustrate that it is possible. One
crucial feature is that there must be a separation of scales - the
dynamics of the system (optimisation, or whatever) must occur more
rapidly than the change to the state space. Otherwise, you get what is
known in the ALife world as a mutational meltdown - evolution ceases
to operate.

--

----------------------------------------------------------------------------
Prof Russell Standish Phone 0425 253119 (mobile)
Mathematics
UNSW SYDNEY 2052 [hidden email]
Australia http://www.hpcoders.com.au
----------------------------------------------------------------------------

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

--
Doug Roberts
[hidden email]
[hidden email]
505-455-7333 - Office
505-670-8195 - Cell

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Steve Smith

Re: Agents, stocks, and flows

In reply to this post by Douglas Roberts-2

Doug -

I think you suggested specifying and designing a system from requirements and something like first principles (which I think is a really good way to approach actually solving the problem).

I suggested a review of the *real world* (not computer systems and models) for networks of entities which are self-modifying in the sense of Russ's quest. I think the answer is (probably) that there are many if we admit long enough time scales (referencing Marcus' comment, the time scale of the evolution/modification of the network itself has to be much longer than the dynamics of the network to avoid "evolutionary meltdown).

I've got an ice-chest full of beers and melting ice if you have time to tip one later today.

- Steve

On Sat, Aug 29, 2009 at 10:53 AM, Steve Smith <[hidden email]> wrote:

I wonder if the combined genius, memory, attention, focus of this group can come up with real-world systems which (obviously) seem to work this way (require this level of abstraction/complexity to model).

-Steve

I thought I suggested this yesterday...

"Has anybody just tried to design this application the old-fashioned way; i.e., develop a set of requirements that

define the interactions between the components of the system,

identify (clearly, no vagueness allowed) the desired results from running the simulation,

identify (clearly, no vagueness allowed) the inputs for the simulation, and

*then* determine what design best fits the application?

[Blah Blah Blah]"

--Doug
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Douglas Roberts-2

Re: Agents, stocks, and flows

That sounds really good. When would be a good time?

On Sat, Aug 29, 2009 at 11:14 AM, Steve Smith <[hidden email]> wrote:

I've got an ice-chest full of beers and melting ice if you have time to tip one later today.

- Steve

Nick Thompson

Re: Agents, stocks, and flows

In reply to this post by Russ Abbott

Steve,

funny, I have an icechest like that, too. Except that mine is in masschusetts.

Notice that there is that that odd copula "self modifying"; I suppose I should get my own damned thread, but some time, I hope we can discuss that idea because, like "self-destroying concept", I think "self-modifying" agent is a self-destroying concept". Or perhaps it's just an oxymoron.

Nicholas S. Thompson

Emeritus Professor of Psychology and Ethology,

Clark University ([hidden email])

http://home.earthlink.net/~nickthompson/naturaldesigns/

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: 8/29/2009 11:16:14 AM

Subject: Re: [FRIAM] Agents, stocks, and flows

Doug -

I think you suggested specifying and designing a system from requirements and something like first principles (which I think is a really good way to approach actually solving the problem).

I suggested a review of the *real world* (not computer systems and models) for networks of entities which are self-modifying in the sense of Russ's quest. I think the answer is (probably) that there are many if we admit long enough time scales (referencing Marcus' comment, the time scale of the evolution/modification of the network itself has to be much longer than the dynamics of the network to avoid "evolutionary meltdown).

I've got an ice-chest full of beers and melting ice if you have time to tip one later today.

- Steve

On Sat, Aug 29, 2009 at 10:53 AM, Steve Smith <[hidden email]> wrote:

I wonder if the combined genius, memory, attention, focus of this group can come up with real-world systems which (obviously) seem to work this way (require this level of abstraction/complexity to model).

-Steve

I thought I suggested this yesterday...

"Has anybody just tried to design this application the old-fashioned way; i.e., develop a set of requirements that

define the interactions between the components of the system,
identify (clearly, no vagueness allowed) the desired results from running the simulation,

identify (clearly, no vagueness allowed) the inputs for the simulation, and

*then* determine what design best fits the application?

[Blah Blah Blah]"

--Doug
============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org

Russ Abbott

Re: Agents, stocks, and flows

I think someone already said this, but here's an easy way to make self-modifying go down easier. Think of an agent that has a built-in genetic programming system. It uses that system to generate a better set of rules for itself, which it then adopts.

That seems both a lot like what we as people do and not so far out as an agent-based modeling framework.

That's essentially the requirement -- or at least one of my requirementss. We are going to do our best to build it -- probably the old fashioned way, in Java.

-- Russ

On Sat, Aug 29, 2009 at 10:36 AM, Nicholas Thompson <[hidden email]> wrote:

Steve,

funny, I have an icechest like that, too. Except that mine is in masschusetts.

Notice that there is that that odd copula "self modifying"; I suppose I should get my own damned thread, but some time, I hope we can discuss that idea because, like "self-destroying concept", I think "self-modifying" agent is a self-destroying concept". Or perhaps it's just an oxymoron.

Nicholas S. Thompson

Emeritus Professor of Psychology and Ethology,

Clark University ([hidden email])

http://home.earthlink.net/~nickthompson/naturaldesigns/
----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: 8/29/2009 11:16:14 AM

Subject: Re: [FRIAM] Agents, stocks, and flows

Doug -

I think you suggested specifying and designing a system from requirements and something like first principles (which I think is a really good way to approach actually solving the problem).

I suggested a review of the *real world* (not computer systems and models) for networks of entities which are self-modifying in the sense of Russ's quest. I think the answer is (probably) that there are many if we admit long enough time scales (referencing Marcus' comment, the time scale of the evolution/modification of the network itself has to be much longer than the dynamics of the network to avoid "evolutionary meltdown).

I've got an ice-chest full of beers and melting ice if you have time to tip one later today.

- Steve

On Sat, Aug 29, 2009 at 10:53 AM, Steve Smith <[hidden email]> wrote:

I wonder if the combined genius, memory, attention, focus of this group can come up with real-world systems which (obviously) seem to work this way (require this level of abstraction/complexity to model).

-Steve

I thought I suggested this yesterday...

"Has anybody just tried to design this application the old-fashioned way; i.e., develop a set of requirements that

define the interactions between the components of the system,
identify (clearly, no vagueness allowed) the desired results from running the simulation,

identify (clearly, no vagueness allowed) the inputs for the simulation, and

*then* determine what design best fits the application?

[Blah Blah Blah]"

--Doug
============================================================ FRIAM Applied Complexity Group listserv Meets Fridays 9a-11:30 at cafe at St. John's College lectures, archives, unsubscribe, maps at http://www.friam.org
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org

Douglas Roberts-2

Re: Agents, stocks, and flows

In reply to this post by Douglas Roberts-2

Shoot. That last one wasn't meant for public consumption. Sorry for sounding like such a poopy-head.

On Sat, Aug 29, 2009 at 11:12 AM, Douglas Roberts <[hidden email]> wrote:

Steve,

I confess that I enjoy rubbing their noses in their speculative circle-jerks, now and then.

Another topic: do you know Ed MacKerrow?

--Doug
On Sat, Aug 29, 2009 at 11:06 AM, Steve Smith <[hidden email]> wrote:
Doug -

Suzanne and I just watched Paper Chase (1973) again and were treated to the dry commentary of John Hausman as curmudgeonly professor Kingsfield.    Your comment reminded me of the equally dry delivery he gave in a TV commercial several years later for an investment house? where he uttered the line

    "We make money the old-fashioned way, we EARN it!"

That noted, I would submit that the FRIAM list (and Friday kaffe klatch) *IS* more about speculative discussion than it is about problem solving. I find there is a delicate distinction between the interesting exploration of ideas through group speculation and well... what I think we all know of as a "circle jerk". I take this risk every time I let myself get drawn into a thread.

- Steve
Has anybody just tried to design this application the old-fashioned way; i.e., develop a set of requirements that

define the interactions between the components of the system,

identify (clearly, no vagueness allowed) the desired results from running the simulation,

identify (clearly, no vagueness allowed) the inputs for the simulation, and

*then* determine what design best fits the application?

Just asking, 'cause this thread so far sniffs out suspiciously like another "I want to talk about how *I* want to think about how (in the purest theoretical sense) simulations should be designed/implemented/thought of."

Just asking...

--Doug

As compared to endlessly seeing that special, elegant simulation implementation system that is the ideal match to this particular problem domain?

On Fri, Aug 28, 2009 at 6:26 PM, russell standish <[hidden email]> wrote:

On Fri, Aug 28, 2009 at 12:46:26PM -0500, Roger Critchlow wrote:
> I don't think you'll find this because it implies programming a higher
> purpose and allowing the agents to jump the rails, as it were, and start
> negotiating their way through the combinatorics of alternative networks.
> Similarly, you won't find models in which agents invent new inputs to
> monitor, new outputs to generate, and new rules which involve new inputs and
> new outputs.
>
> Optimization within a fixed solution space, which is what we do when we let
> agents play with the flow through a fixed network or let them search out the
> most profitable rules in a set of prespecified alternatives, gets hairy
> enough without opening things up to the infinity of potential solutions that
> we didn't have time to program into the model ourselves.
>

EcoLab is an example of a model where the state space evolves over
time rather than staying fixed. It is not quite the ABM that Russ
Abbott is looking for, but does illustrate that it is possible. One
crucial feature is that there must be a separation of scales - the
dynamics of the system (optimisation, or whatever) must occur more
rapidly than the change to the state space. Otherwise, you get what is
known in the ALife world as a mutational meltdown - evolution ceases
to operate.

--

----------------------------------------------------------------------------
Prof Russell Standish Phone 0425 253119 (mobile)
Mathematics
UNSW SYDNEY 2052 [hidden email]
Australia http://www.hpcoders.com.au
----------------------------------------------------------------------------

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
lectures, archives, unsubscribe, maps at http://www.friam.org
--
Doug Roberts
[hidden email]
[hidden email]
505-455-7333 - Office
505-670-8195 - Cell