Efficient Global Optimization using Deep Gaussian Processes

TL;DR

This paper enhances Efficient Global Optimization by integrating Deep Gaussian Processes to better handle non-stationary functions, addressing limitations of traditional Gaussian Process models in expensive black-box optimization.

Contribution

It introduces the use of Deep Gaussian Processes within EGO to improve modeling of non-stationary functions, exploring associated challenges and benefits.

Findings

DGP improves modeling of non-stationary functions in EGO

Numerical experiments demonstrate enhanced optimization performance

Identifies challenges in integrating DGP with EGO

Abstract

Efficient Global Optimization (EGO) is widely used for the optimization of computationally expensive black-box functions. It uses a surrogate modeling technique based on Gaussian Processes (Kriging). However, due to the use of a stationary covariance, Kriging is not well suited for approximating non stationary functions. This paper explores the integration of Deep Gaussian processes (DGP) in EGO framework to deal with the non-stationary issues and investigates the induced challenges and opportunities. Numerical experimentations are performed on analytical problems to highlight the different aspects of DGP and EGO.