Learn R Programming

activegp (version 1.1.1)

Gaussian Process Based Design and Analysis for the Active Subspace Method

Description

The active subspace method is a sensitivity analysis technique that finds important linear combinations of input variables for a simulator. This package provides functions allowing estimation of the active subspace without gradient information using Gaussian processes as well as sequential experimental design tools to minimize the amount of data required to do so. Implements Wycoff et al. (JCGS, 2021) .

Copy Link

Version

Install

install.packages('activegp')

Monthly Downloads

274

Version

1.1.1

License

BSD_3_clause + file LICENSE

Maintainer

Nathan Wycoff

Last Published

May 25th, 2024

Functions in activegp (1.1.1)

Covariance of kernel computations

Covariance of kernel computations

C update with new observations

Get the distance between subspaces defined as the ranges of A and B

Covariance of kernel computations

f:[-1, 1] -> R Becomes f:[0,1] -> R

Plot const_C objectc

grad_est_subspace

Estimate the Active Subspace of a Cheap Function using Gradients

Update Constantine's C, using update formula

Hessian of the log-likelihood with respect to lengthscales hyperparameters Works for homGP and hetGP models from the hetGP package for now.

Element-wise Cn+1 variance

Expected variance of trace of C

Equivalent of C_GP using RcppArmadillo

Active Subspace Matrix closed form expression for a GP.

Alternative Variance of Update

Active Subspace Prewarping

CI on Eigenvalues via Monte Carlo/GP

as.matrix.const_C

Package activegp

Active subspace for second order linear model

Covariance of kernel computations

Change a function's inputs to live in [-1, 1]

grad_W_kappa_ij2_w2

Covariance of kernel computations

Quantities for Acquisition Functions

grad_W_kappa_ij2

Covariance of kernel computations

Get Integerodifferential Quantities

Covariance of kernel computations

Rectangular Domain -> Unbounded Domain

Print const_C objects