Structured Deep Kernel Networks for Data-Driven Closure Terms of Turbulent Flows

TL;DR

This paper introduces an extension of Structured Deep Kernel Networks (SDKN) that effectively handles large, high-dimensional datasets and outperforms neural networks in predicting closure terms of turbulent flows.

Contribution

The paper presents an extended SDKN approach that combines with standard ML modules and demonstrates superior performance in data-driven turbulence modeling.

Findings

SDKN handles large datasets effectively

Achieves near-perfect accuracy in turbulence closure prediction

Outperforms neural networks in the given application

Abstract

Standard kernel methods for machine learning usually struggle when dealing with large datasets. We review a recently introduced Structured Deep Kernel Network (SDKN) approach that is capable of dealing with high-dimensional and huge datasets - and enjoys typical standard machine learning approximation properties. We extend the SDKN to combine it with standard machine learning modules and compare it with Neural Networks on the scientific challenge of data-driven prediction of closure terms of turbulent flows. We show experimentally that the SDKNs are capable of dealing with large datasets and achieve near-perfect accuracy on the given application.