4 ms·
John von Neumann and stochastic simulations
- claytonjy 11y agoWhat's the current state of the "stochasticity" view presented here? I've always thought of Monte Carlo as being a convenient way to avoid expensive analytical solutions, rather than a reproduction of a stochastic reality, but I'm not a frequent user of the techniques.
- jcfrei 11y agoDepending on the quality of your model I think it can be very much a good approximation of "stochastic reality" (ie. the real world probabilities).
- p4wnc6 11y agoFor more on the view that stochasticity fundamentally parallels reality, an interesting read is "The Dawning of the Age of Stochasticity" by David Mumford [1]. I also think it's interesting that methods like Hybrid Monte Carlo were inspired by a physical-reality-first idea. It makes it seem like this is somewhat of a chicken and egg problem. [1] http://www.stat.uchicago.edu/~lekheng/courses/191f09/mumford-AMS.pdf http://www.stat.uchicago.edu/~lekheng/courses/191f09/mumford...