Efficient Global Optimization using Deep Gaussian Processes

2018/09/11 by Ali Hebbal, Hebbal, Ali, Loic Brevault +7 · 1 citation
Computer Science · Mathematics · #FOS: Computer and information sciences #FOS: Mathematics #Machine Learning (cs.LG) #Neural and Evolutionary Computing (cs.NE) #Optimization and Control (math.OC) #cs.LG #cs.NE #math.OC

paper · pdf · doi:10.48550/arxiv.1809.04632

12 pages, 11 Figures, 2 Tables, presented to the IEEE Congress on Evolutionary Computation (IEEE CEC 2018)

arxiv created 2018/09/11 · arxiv updated 2018/09/14

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.

Cited by

• GRIMIP: A General Framework for Instance-Specific Configuration of MIP Solvers Using LLMs

Related

• Small ensembles of kriging models for optimization
• Bayesian Optimization using Deep Gaussian Processes
• Hyper-optimization with Gaussian Process and Differential Evolution Algorithm
• The True Destination of EGO is Multi-local Optimization
• Combined Global and Local Search for Optimization with Gaussian Process Models