Sensitivity-Aware Surrogate Modeling with Random Feature Expansions
Organizations: Dept. of Mathematics, Emory University, Atlanta, GA 30322, USA
Abstract
Random feature expansions (RFEs) provide a flexible approach to surrogate modeling, but their accuracy depends on the distribution used to sample features. We propose spike-and-slab random feature expansions (SAS-RFE) to learn the feature sampling distribution from sensitivity analysis. The method employs a spike-and-slab distribution that assigns probabilities to the inclusion of individual inputs in randomly sampled Fourier features. These inclusion probabilities are determined by estimates of total Sobol' sensitivity indices, rather than by explicitly selecting important inputs or interaction terms. An information theoretic criterion shows that the total Sobol' indices yield a distribution that approximates the input interaction structure of the model. An iterative algorithm updates the sampling probabilities using sensitivity estimates from successive RFE surrogates. Analytical formulas for the Sobol' indices of random Fourier feature expansions enable these updates without additional evaluations of the target function. Numerical experiments on six benchmark problems, with up to 100 input dimensions, demonstrate improvements in surrogate approximation and sensitivity estimation.
Figures & tables
| Input | Type | Parameters | Mean | St. Dev. |
|---|---|---|---|---|
| Lognormal | ||||
| Lognormal | ||||
| Lognormal | ||||
| – | Gumbel |
| Input | Description | Type | Parameters |
|---|---|---|---|
| Radius of borehole | Gaussian | ||
| Radius of influence | Lognormal | ||
| Transmissivity of upper aquifer | Uniform | ||
| Potentiometric head of upper aquifer | Uniform | ||
| Transmissivity of lower aquifer | Uniform | ||
| Potentiometric head of lower aquifer | Uniform |
| Input | Description | Type | Parameters |
|---|---|---|---|
| Fraction of absorption ( | Weibull | ||
| Absorption rate | Lognormal | ||
| Gut availability ( ) | Weibull | ||
| Blood to plasma ratio (BP) | Gaussian | ||
| Fraction of drug unbound in plasma | Lognormal | ||
| Liver tissue to plasma partition coefficient ( ) | Gaussian |
| Model | Dense RFE | SAS-RFE | PCE |
|---|---|---|---|
| g-func. | 1.39 | 0.607 | 0.545 |
| O. & O. | 0.334 | ||
| 100D | |||
| Truss | |||
| Borehole | |||
| PBPK |
| Model | Dense RFE | SAS-RFE | PCE |
|---|---|---|---|
| g-func. | |||
| O. & O. | |||
| 100D | |||
| Truss | |||
| Borehole | |||
| PBPK |
Surrogate Modeling
Surrogate modeling builds faster or more interpretable approximations of expensive simulations and black-box models.
See all 253 papers