Busby, Daniel. "Hierarchical adaptive experimental design for Gaussian process emulators." Reliability Engineering and System Safety. 94 (2009): 1183-1193. Electronic.
Researching computer simulations.
"Computer models, denoted as simulators, are used to predict the behavior of complex physical systems." However, the more complex the situation, the longer the simulations take. Using Bayesian statistics, there are algorithms to predict the "best" area to sample or simulate to allow for the most useful collection of data points. This paper proposed an new design called hierarchical adaptive design that uses the least possible number of simulations. The most useful part of this article is the structure. Basically, this will be the structure of what my thinking should be.
No comments:
Post a Comment