Surrogate modeling

From Rave Documentation
Revision as of 17:32, 22 August 2013 by Admin (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Surrogate modeling refers to the process of creating a function from data that approximates the relation underlying the data. For example if you have a data set with two variables, Height and Weight, you could create a surrogate model Weight = f(Height) so that given you data for Height, you can generate corresponding values of Weight.

Surrogate modeling is used for:

  • Enabling capabilities that depend on functional variables, such as optimization and continuous visualization
  • Speeding up calculations: surrogate models are generally fast, relatively simple mathematical expressions that can be evaluated almost instantly.
  • And more...

Rave can create many types of surrogate models via an easy-to-use interface. Please note that although Rave tries to make the modeling process as user-friendly as possible, significant expertise is required to create good surrogate models.