Gradient of Gaussian Process Regressor

Asked Jun 10 '23 at 05:29

Active Jun 10 '23 at 06:39

Viewed 58 times

I have a data ((x,y),f) that I am fitting using Gaussian Process Regression in Python's sklearn package. The posterior mean of the GP is essentially my output with an associated error. Based on either the posterior mean or the original data itself, is there a way to calculate gradients of the original data via GP? x and y are features. This is not a 1D problem. Can someone suggest me a 2D solution.

I am trying to solve a 2D Non-Linear Poisson's Equation. I need the values of both prediction and gradient to solve this.

edited Jun 10 '23 at 06:39

asked Jun 10 '23 at 05:29

Prince SBI

What do you need the gradients for? – Tim Jun 10 '23 at 05:59
1

I am trying to solve a 2D Non-Linear Poisson's Equation. I need the values of both prediction and gradient to solve this. – Prince SBI Jun 10 '23 at 06:38
check section 9.4 of R&W (https://gaussianprocess.org/gpml/chapters/RW.pdf) for a very brief description – John Madden Jun 10 '23 at 15:45
Related/answered here – g g Jun 10 '23 at 18:23

Gradient of Gaussian Process Regressor

0 Answers0