Isotonic Surrogate Modeling for Computer Experiments with Many Input Variables

License: See LICENSE file · Updated 2026-09-28
My rating

The paper addresses the computational cost of virtual simulators used to study complex physical phenomena such as particle collisions and rocket propulsion. It focuses on Bayesian surrogate models for computer experiments with many input variables and limited sample sizes. The approach incorporates monotonicity or isotonicity information that can be elicited from physical systems to support more accurate and uncertainty-aware emulation.

Tags: