[Vp-reproduce-subgroup] [Vp-integration-subgroup] [EXT] Re: White paper revision

[Jacob Barhak]

Yes William,

Winston is right, there are multiple types of operator and those extend
beyond math definition. In fact ensemble models use multiple compositions
methods - your own simple example that you shown a few week ago on this
mailing list needs to be extended to many more examples - I think
composition methods are a topic for a paper on its own with many examples.
For now allow me to be happy to see this conversation.

             Jacob

On Tue, May 18, 2021 at 3:19 PM William Waites <wwaites at ieee.org> wrote:

> Winston, you have hit the nail on the head! Maybe we would want to think
> of two basic kinds of combination operator, loosely, one that composes in
> models series and one that does this in parallel. And maybe think of these
> as a coproduct and product in a monoidal category. As you say, specifying
> the object of that category precisely, and defining these operations
> properly to be able to get an algebra of models (for at least some kinds of
> model, and we do actually have this for some kinds) is the tricky bit.
>
> I’d also like to contribute to #9 and also the one about stochastic models.
>
> Cheers,
> -w
>
>
> > On 18 May 2021, at 21:01, Winston Garira <Winston.Garira at univen.ac.za>
> wrote:
> >
> > I could not attend the zoom meeting because there was another meeting
> with graduate students at  the same time.  I have read through the
> interesting communications that have been ongoing since yesterday from the
> subgroup. I will be able to contribute and edit challenge 9:  "Environments
> to adapt and integrate models". Already an interesting discussion has  been
> initiated in the form of  the following statements:
> >
> > >If you separately accredit M1 and M2 for your purpose, and they fit
> together like lego, it does not >automatically mean that M1 + M2 is
> credible. One of the things we would like to know is, what >properties must
> M1 and M2 have for us to safely believe that we understand what M1 + M2
> does. We >only know this for a few special cases today.
> >
> > Two issues need consideration here. If M1and M2 are different models
> describing phenomena at the same scale, then perhaps this is the few
> special cases where we may know that  M1+M2 is credible. This is because
> “+” is much easier to define in that case. However, if M1 and M2 describe
> phenomena at different scales then  the problem of  defining  “+” becomes
> even more challenging. Defining “+” involves defining methods for
> linking/coupling/integrating  models from the different  scales.   For
> multiscale modelling, the different  scales  indicate/represent
> shifts/changes/transitions in  processes for the complex system under
> consideration and we do not know in general how scale transition occurs.
> For example in infectious disease dynamics there is usually pathogen
> replication at the microscale scale and pathogen  transmission at the
> macroscale [1].  There is no single unique  and widely acceptable
> method/way of defining “+” or integrating models for a given M1 and M2.  In
> the past, I have considered the different ways in which to define “+”
>  when M1 and M2 describe infectious disease phenomena at different scales
> (i.e the microscale and macroscale). This enabled me to come up with 5
> different categories of multiscale  models of disease dynamics [2,3].  So
> even if  M1 and M2  have the correct properties we may not safely  believe
> what M1 + M2 does because there  is still an additional  challenge of
> identifying  the correct “+” which turns out to be the greatest challenge
> in multiscale modelling.
> >
> > [1.] Garira, W. (2019). The Replication-Transmission Relativity theory
> for Multiscale Modelling of infectious Disease Systems. Scientific reports,
> 9(1), 1-17.
> > [2.] Garira, W. (2017). A complete categorization of multiscale models
> of infectious disease systems. Journal of biological dynamics, 11(1),
> 378-435.
> > [3.] Garira, W. (2018). A primer on multiscale modelling of infectious
> disease systems. Infectious Disease Modelling, 3, 176-191.
> >
> > Winston Garira, PhD
> > DEPARTMENT OF MATHEMATICS AND APPLIED MATHEMATICS
> > Modelling Health and Environmental Linkages Research Group (MHELRG)
> > University of Venda, Private Bag X5050, Thohoyandou 0950, South Africa.
> > Email(1): winston.garira at univen.ac.za, Email(2): wgarira at gmail.com
> >
> > From: Jacob Barhak <jacob.barhak at gmail.com>
> > Sent: Tuesday, 18 May 2021 17:17
> > To: William Waites <wwaites at ieee.org>
> > Cc: Tomas Helikar <thelikar2 at unl.edu>;
> vp-reproduce-subgroup at lists.simtk.org <
> vp-reproduce-subgroup at lists.simtk.org>;
> vp-integration-subgroup at lists.simtk.org <
> vp-integration-subgroup at lists.simtk.org>; Winston Garira <
> Winston.Garira at univen.ac.za>; Faeder, James R <faeder at pitt.edu>
> > Subject: Re: [Vp-integration-subgroup] [EXT] Re: White paper revision
> >
> > So William,
> >
> > For now, you should just claim the sections you wish to edit - I think
> your name is mentioned already, yet check out the rest of the sections.
> This will give John a picture of who he has to satisfy when he makes edits.
> >
> > Once he has a full picture we can start actual editing. He will decide
> where edits go - I suggested using the existing version so changes are
> reflected in one location and we will have a traceable track of changes -
> if he prefers another link it's fine - yet I asked him to secure the
> document and provide links to older versions the document is derived from.
> It is after all a paper about reproducibility. And I did experience
> security issues in the past in collaborative work that needed treatment, so
> this is why I asked for those. You really do not want an anonymous animal
> using your public document as it sees fit - remember that our conversations
> are public as well as the links.
> >
> > He promised 2 weeks and I do not envy him if he has to go through the
> history of contributions and locate authors who contributed to each
> section, this is why we are declaring what sections are are interested in -
> Sheriff suggested this yesterday. So please just look at the sections John
> mentioned with his numbering and declare what you want to edit. Once we all
> claimed sections, John can more easily decide what to edit and how without
> facing potential objections.
> >
> > I hope this makes sense now and you know your next move.
> >
> >           Jacob
> >
> >
> >
> >
> >
> >
> > On Tue, May 18, 2021 at 9:57 AM William Waites <wwaites at ieee.org> wrote:
> >
> > > On 18 May 2021, at 15:30, Tomas Helikar <thelikar2 at unl.edu> wrote:
> > >
> > > Since John is taking the lead on the new version, my suggestion is to
> let him do it the way that will be most efficient for him. I think
> everyone, including John, have noted your concerns.
> >
> > I agree. But I am also now completely confused about which document it
> is ok for who to edit when.
> >
> > -w
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.simtk.org/pipermail/vp-reproduce-subgroup/attachments/20210518/8a0dfad5/attachment.html>