how to build trust in computer simulations: Towards a general epistemology of validation

I have rarely attended a workshop with such a precise goal, but then I have neither ever attended a philosophy workshop… Tonight, I am flying to Han(n)over, Lower Saxony, for a workshop on the philosophical aspects of simulated models. I was quite surprised to get invited to this workshop, but found it quite a treat to attend a multi-disciplinary meeting about simulations and their connection with the real world! I am less certain I can contribute anything meaningful, but still look forward to it. And will report on the discussions, hopefully. Here is the general motivation of the workshop:

“In the last decades, our capacities to investigate complex systems of various scales have been greatly enhanced by the method of computer simulation. This progress is not without a price though: We can only trust the results of computer simulations if they have been properly validated, i.e., if they have been shown to be reliable. Despite its importance, validation is often still neglected in practice and only poorly understood from a theoretical perspective. The aim of this conference is to discuss methodological and philosophical problems of validation from a multidisciplinary perspective and to take first steps in developing a general framework for thinking about validation. Working scientists from various natural and social sciences and philosophers of science join forces to make progress in understanding the epistemology of validation.”

3 Responses to “how to build trust in computer simulations: Towards a general epistemology of validation”

  1. Hi,
    In my experiences as researcher I think one problem is the positive assumption to simulate what one expect. You get always the results you want out of a simulation.

    I would like to hear more and I hope you can summarize the conference for us. I also interested in a reference list.

    Marcus Landschulze

    • This was my summary, de facto, and no reference list was provided during or after the workshop. I suggest you check against all participants and their webpage, to see their list of publication. As I did during the conference, hence the links to some books by some speakers… Good luck!

  2. I think that (statistical) simulation is a bit of a generalization of math (specifically what Stigler referred to a Euclidean math) which can be nicely expressed using Peirce’s diagrammatic definition of math as experiments conducted on diagrams (or more generally symbols) instead of empirical objects (e.g. chemicals). If the experiment described is in fact valid others can re-run them and get the same result by doing the same manipulation on the same materials.

    In math you get the same exactly the same result where as in simulation (for valid simulations) you get results that are just equal in distribution. This doesseem to make it not the usual kind of math.

    As for how abstraction math relates to (represents) reality I don’t see how there would be a difference?

    Enjoy the conference, hopefully you can summarize for us!

    Keith O’Rourke

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: