[SYSTEMDS-1994] Implementation of Gaussian Process Regression

• Type: Sub-task
• Status: In Progress
• Priority: Major
• Resolution: Unresolved
• Fix Version/s: SystemML 1.1
• Component/s: Algorithms
• Assignee: Janardhan Pulivarthi
• Reporter: Janardhan Pulivarthi
• Created: 06/Nov/17 14:26
• Updated: 17/Apr/20 11:02

Description

Regression with Gaussian process.

This script essentially takes in the input  ( X, y ) (input is in matrix format), then 

1. It calculates a covariance matrix,  K
2. and a predictive mean and variance of a test point x_star (it is a single point).
3. a log marginal likelihood

Algorithm : taken from gaussian processes for machine learning by Williams & Rasmussen (page 19)

Issue Links

• links to: GitHub Pull Request #731