One of my projects
Locked
 
|
All,
Here is a press release on one of the projects that I am working on at RTI International. We are doing this work under a grant from the NFS. Fun stuff. http://www.rti.org/page.cfm?nav=640&objectid=E441EE2F-04EB-4593-A2C3F6930685E110 --Doug -- Doug Roberts, RTI International droberts at rti.org doug at parrot-farm.net 505-455-7333 - Office 505-670-8195 - Cell -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070328/4c24c96d/attachment.html |
|
|
Congratulations, Doug. Looks very cool!
> -----Original Message----- > From: Douglas Roberts [mailto:doug at parrot-farm.net] > Sent: Wednesday, March 28, 2007 5:28 PM > To: The Friday Morning Applied Complexity Coffee Group > Subject: [FRIAM] One of my projects > > All, > > Here is a press release on one of the projects that I am > working on at RTI International. We are doing this work > under a grant from the NFS. Fun stuff. > > http://www.rti.org/page.cfm?nav=640&objectid=E441EE2F-04EB-459 > 3-A2C3F6930685E110 > <http://www.rti.org/page.cfm?nav=640&objectid=E441EE2F-04EB-45 > 93-A2C3F6930685E110> > > --Doug > > > -- > Doug Roberts, RTI International > droberts at rti.org > doug at parrot-farm.net > 505-455-7333 - Office > 505-670-8195 - Cell > |
|
|
Thanks, Stephen.
The difference between the depth of description given a technical project in a press release and the actual depth of description that would be required to fully explain the implementation details, is like, well, the difference between 1. night and day, 2. black/white, 3. some of the discussions of 'agent-based simulations' that time and again occur on FRIAM. Present company excluded, of course. ;-} Suffice it to say that there are some butt-ugly nuts & bolts that will underlie this particular initiative. --Doug -- Doug Roberts, RTI International droberts at rti.org doug at parrot-farm.net 505-455-7333 - Office --Doug On 3/28/07, Stephen Guerin <stephen.guerin at redfish.com> wrote: > > Congratulations, Doug. Looks very cool! > > > -----Original Message----- > > From: Douglas Roberts [mailto:doug at parrot-farm.net] > > Sent: Wednesday, March 28, 2007 5:28 PM > > To: The Friday Morning Applied Complexity Coffee Group > > Subject: [FRIAM] One of my projects > > > > All, > > > > Here is a press release on one of the projects that I am > > working on at RTI International. We are doing this work > > under a grant from the NFS. Fun stuff. > > > > http://www.rti.org/page.cfm?nav=640&objectid=E441EE2F-04EB-459 > > 3-A2C3F6930685E110 > > <http://www.rti.org/page.cfm?nav=640&objectid=E441EE2F-04EB-45 > > 93-A2C3F6930685E110> > > > > --Doug > > > > > > -- > > Doug Roberts, RTI International > > droberts at rti.org > > doug at parrot-farm.net > > 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 > 505-670-8195 - Cell -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070328/50e5c01a/attachment.html |
|
|
In reply to this post by Douglas Roberts-2
Doug, this is way cool. I went to the TeraGrid site to see if there was any info on how the nine sites are connected among themselves and to users, but couldn't find anything. Can you sketch the telecom infrastructure and protocol being used to manage this distributed resource?
db dba | David Breecker Associates, Inc. www.BreeckerAssociates.com Abiquiu: 505-685-4891 Santa Fe: 505-690-2335 ----- Original Message ----- From: Douglas Roberts To: The Friday Morning Applied Complexity Coffee Group Sent: Wednesday, March 28, 2007 4:27 PM Subject: [FRIAM] One of my projects All, Here is a press release on one of the projects that I am working on at RTI International. We are doing this work under a grant from the NFS. Fun stuff. http://www.rti.org/page.cfm?nav=640&objectid=E441EE2F-04EB-4593-A2C3F6930685E110 --Doug -- Doug Roberts, RTI International droberts at rti.org doug at parrot-farm.net 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 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070328/99fdeaab/attachment.html |
|
|
David Breecker wrote:
> I went to the TeraGrid site to see if there was any info on how the > nine sites are connected among themselves and to users, but couldn't > find anything. Can you sketch the telecom infrastructure and protocol > being used to manage this distributed resource? Fiber optic at 10 gigabit/sec and above: http://en.wikipedia.org/wiki/Lambda_rail_network http://en.wikipedia.org/wiki/Internet2 Software includes typical HPC packages, but also: http://www.globus.org http://www-03.ibm.com/systems/clusters/software/gpfs.html |
|
|
An HTML attachment was scrubbed...
URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070329/535dd995/attachment.html |
|
|
In my role as FRIAM's official Cassandra (I should get a T-shirt printed),
has anyone ever shown that these highly intensive simulations give quantitatively better results than, say, something written on Owen's laptop in NetLogo? Do we know that we get a better assessment of (for example) the robustness of policies for stopping epidemic spread or do we rely on the "more is better" argument? ("Of course, the results are better - we have an NSF grant and 15 supercomputers"). R -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070329/38c97c3c/attachment.html |
|
|
Shhh!
Someone might hear you! Seriously, good question, Robert. The answer, of course, is that it depends on the questions being asked about the system being simulated. If the questions are such that a simple simulation can provide answers, then force-fitting a large, more detailed simulation to provide results will probably be of no advantage whatever. It is for systems -- typically highly complex, dynamic systems -- when certain questions are asked for which simple simulations can provide no relevant insight that a larger, more complex simulation can sometimes effectively be brought to bear. All the usual codicils apply to the above paragraph, such as 1. the big, detailed simulation is detailed in the proper areas to address the analysis requirements, 2. the analysis requirements make sense, 3. data exists to support the detailed rendering of the physical system being modeled, and 4. the analyst knows what he's doing. The reality is that there is no one-size-fits-all solution for simulations. --Doug -- Doug Roberts, RTI International droberts at rti.org doug at parrot-farm.net 505-455-7333 - Office 505-670-8195 - Cell On 3/29/07, Robert Holmes <robert at holmesacosta.com> wrote: > > In my role as FRIAM's official Cassandra (I should get a T-shirt printed), > has anyone ever shown that these highly intensive simulations give > quantitatively better results than, say, something written on Owen's laptop > in NetLogo? Do we know that we get a better assessment of (for example) the > robustness of policies for stopping epidemic spread or do we rely on the > "more is better" argument? ("Of course, the results are better - we have an > NSF grant and 15 supercomputers"). > > R > > ============================================================ > 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 > An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070329/d96cce83/attachment.html |
|
|
In reply to this post by Robert Holmes
Robert Holmes wrote:
> In my role as FRIAM's official Cassandra (I should get a T-shirt > printed), has anyone ever shown that these highly intensive > simulations give quantitatively better results than, say, something > written on Owen's laptop in NetLogo? Do we know that we get a better > assessment of (for example) the robustness of policies for stopping > epidemic spread or do we rely on the "more is better" argument? ("Of > course, the results are better - we have an NSF grant and 15 > supercomputers"). In a model, each added parameter has a statistical cost that needs to be justified statistically in terms of how more accurate the simulation prediction becomes. It happens that for agent models, there's also a strong practical incentive to reduce them because the sensitivity analysis gets so hard to do. For example, a sensitivity analysis can easily involve a huge space to explore even making many simplifying assumptions. If simulation only runs for 1 minute and there are 65536 plausible simulation configurations (e.g. 16 switches to throw, presumably from a space of many more implausible ones), that's still 45 days of runtime on one computer to fully explore the space and build a distribution of outcomes. It is not hard to find 16 plausible parameters to put into a realistic model. Rare sequences of events can be important to disease propagation and their consequences can depend very much on population density, transportation behaviors of that population, and available health interventions. A modeler could describe a particular parameterization of a particular microcosm, but that's not probably what most planners are interested in, except to the extent some are just fishing around for a cartoon dramatization to show their boss. Unfortunately, to observe the rare event in a realistic setting, each instance of a simulation can be pretty substantial, because you're waiting for those rare events in a changing environment. Given a goal to correlate that changing environment with the rare events, it becomes all the harder because the frequently at which one can observe the co-occurring events is even more rare, thus getting a confidence interval on those combinations is hard. Facts about model components can be hard to come by, independent of intuition of dynamics about the whole thing behaves. You may not know exactly how a disease behaves in specific contexts, so you have to cover a range of possibility (e.g. one switch setting being the high and low range of what you think could happen best and worst case, assuming that it is linear and independent of other variables, which it may not be). Marcus |
|
|
In reply to this post by Douglas Roberts-2
This is very, very cool stuff. You should give a WedTech talk on the
butt ugly nuts and bolts. Being someone who studies people who use models, I'm curious about how you guys are relating to your user community. Who are the intended analysts (the ones that you hope know what you're doing)? At what point do you guys start engaging them? Do they treat your simulation as black box? Laura ________________________________ From: [hidden email] [mailto:[hidden email]] On Behalf Of Douglas Roberts Sent: Wednesday, March 28, 2007 5:28 PM To: The Friday Morning Applied Complexity Coffee Group Subject: [FRIAM] One of my projects All, Here is a press release on one of the projects that I am working on at RTI International. We are doing this work under a grant from the NFS. Fun stuff. http://www.rti.org/page.cfm?nav=640&objectid=E441EE2F-04EB-4593-A2C3F693 0685E110 --Doug -- Doug Roberts, RTI International droberts at rti.org doug at parrot-farm.net 505-455-7333 - Office 505-670-8195 - Cell -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070329/06fbb17b/attachment.html |
|
|
Fair enough: big simulation answers some questions, small simulation answers
others. So what are the specific questions that a big epidemiological simulation can answer? It can't be anything too predictive ("ohmigod, New York has just fallen to small pox. Which city is next?") because that depends (I'd guess) on something that is unsimulatable ("errr.... dunno. Kinda depends which flight the guy with small pox got onto"). What are the questions that can only be answered with a big model? R -------------- next part -------------- An HTML attachment was scrubbed... URL: http://redfish.com/pipermail/friam_redfish.com/attachments/20070329/c5ed51d1/attachment.html |
|
|
Robert Holmes wrote:
> So what are the specific questions that a big epidemiological > simulation can answer? It can't be anything too predictive ("ohmigod, > New York has just fallen to small pox. Which city is next?") because > that depends (I'd guess) on something that is unsimulatable ("errr.... > dunno. Kinda depends which flight the guy with small pox got onto"). In addition to regional parameters like average social network size and frequency of interactions by individuals to connected social networks, there are a finite number of flights to get on to, with known flight volumes, and they could all be considered in order to create maps of risk depending on different starting locations given empirical info on transmission conditions.. |
«
Return to Friam
|
1 view|%1 views
| Free forum by Nabble | Edit this page |