.png){kind=icon}
.png){kind=icon}
Return
A mixed-categorical correlation kernel for Gaussian process
.png){kind=icon}
OA
AI
DOI：10.1016/j.neucom.2023.126472.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Recently, there has been a growing interest for mixed-categorical meta-models based on Gaussian pro-cess (GP) surrogates. In this setting, several existing approaches use different strategies either by using continuous kernels (e.g., continuous relaxation and Gower distance based GP) or by using a direct estima-tion of the correlation matrix. In this paper, we present a kernel-based approach that extends continuous exponential kernels to handle mixed-categorical variables. The proposed kernel leads to a new GP surro-gate that generalizes both the continuous relaxation and the Gower distance based GP models. We demonstrate, on both analytical and engineering problems, that our proposed GP model gives a higher likelihood and a smaller residual error than the other kernel-based state-of-the-art models. Our method is available in the open-source software SMT.& COPY; 2023 Elsevier B.V. All rights reserved.
Keywords:
Gaussian process
Mixed-categorical
Continuous relaxation
Hypersphere decomposition
Bayesian optimization
Surrogate modeling toolbox
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.5
Papers:
2.5W
Citations:
6.5W
View more organizations