integrand_edm_crit

Computes the integrand of the distance in measure criterion.

Computes pseudo-realizations from the posterior distribution of a Gaussian Process (GP) with the method described in Azzimonti et al. (2016) <doi:10.1137/141000749>. The realizations are obtained from simulations of the field at few well chosen points that minimize the expected distance in measure between the true excursion set of the field and the approximate one. Also implements a R interface for (the main function of) Distance Transform of sampled Functions (<https://cs.brown.edu/people/pfelzens/dt/index.html>).

Dario Azzimonti

pGPx

Pseudo-Realizations for Gaussian Process Excursions

Julien Bect

Pedro Felzenszwalb

integrand_edm_crit function

<dl><dt>x</dt>
<dd>vector of dimension \(d\) representing the \(ith\) point where to compute the criterion</dd>
<dt>E</dt>
<dd>matrix of dimension \(d*(i-1)\) containing the previously optimized simulation points</dd>
<dt>model</dt>
<dd>km model</dd>
<dt>Thresh</dt>
<dd>threshold selected for excursion set</dd>
<dt>batchsize</dt>
<dd>number of simulation points</dd>
<dt>alpha</dt>
<dd>value of Vorob'ev threshold</dd>
<dt>predE</dt>
<dd>list containing the posterior mean and standard deviation at E</dd>
<dt>predx</dt>
<dd>list containing the posterior mean and standard deviation at x</dd>
<dt>precalc.data</dt>
<dd>list result of precomputeUpdateData with <code>model</code> and <code>x</code>.</dd></dl>

Arguments

Integrand of the distance in measure criterion — integrand_edm_crit

<dl>

<dt>x</dt>
<dd>vector of dimension \(d\) representing the \(ith\) point where to compute the criterion</dd>


<dt>E</dt>
<dd>matrix of dimension \(d*(i-1)\) containing the previously optimized simulation points</dd>


<dt>model</dt>
<dd>km model</dd>


<dt>Thresh</dt>
<dd>threshold selected for excursion set</dd>


<dt>batchsize</dt>
<dd>number of simulation points</dd>


<dt>alpha</dt>
<dd>value of Vorob'ev threshold</dd>


<dt>predE</dt>
<dd>list containing the posterior mean and standard deviation at E</dd>


<dt>predx</dt>
<dd>list containing the posterior mean and standard deviation at x</dd>


<dt>precalc.data</dt>
<dd>list result of precomputeUpdateData with <code>model</code> and <code>x</code>.</dd>

</dl>

integrand_edm_crit: Integrand of the distance in measure criterion

Description

Usage

Value

Arguments

References