Exploring Multi-Fidelity Networks and Adapting their Architecture: A Paradigm for Enhanced Learning and Efficiency

Bayan Hamdan, Pingfeng Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Multi-Fidelity Networks (MFNets) have emerged as a promising approach for surrogate modeling when dealing with limited data and heterogeneous models. They provide a means to establish relationships between different models based on their parameters, rather than relying solely on inputs or outputs. The covariance matrix, which captures the interconnections between the parameters, typically follows a peer structure assumption. However, when the low-fidelity models exhibit dependencies, alternative model architectures can be considered to better capture the underlying relationships. This paper proposes a modified MFNets model that incorporates a hierarchical structure and presents a generalized formulation applicable to diverse applications. A benchmark numerical problem is implemented to demonstrate the advantages of considering different underlying model architectures. The results showcase improved predictive capabilities of MFNets when estimating high-fidelity functions.

Original languageEnglish (US)
Title of host publication50th International Conference on Computers and Industrial Engineering, CIE 2023
Subtitle of host publicationSustainable Digital Transformation
EditorsYasser Dessouky, Abdulrahim Shamayleh
PublisherComputers and Industrial Engineering
Pages888-896
Number of pages9
ISBN (Electronic)9781713886952
StatePublished - 2023
Event50th International Conference on Computers and Industrial Engineering: Sustainable Digital Transformation, CIE 2023 - Sharjah, United Arab Emirates
Duration: Oct 30 2023Nov 2 2023

Publication series

NameProceedings of International Conference on Computers and Industrial Engineering, CIE
Volume2
ISSN (Electronic)2164-8689

Conference

Conference50th International Conference on Computers and Industrial Engineering: Sustainable Digital Transformation, CIE 2023
Country/TerritoryUnited Arab Emirates
CitySharjah
Period10/30/2311/2/23

Keywords

  • Heterogeneous Models
  • Limited Data
  • Model Architecture
  • Multi-Fidelity Networks
  • Surrogate Modeling

ASJC Scopus subject areas

  • General Computer Science
  • Control and Systems Engineering
  • Electrical and Electronic Engineering
  • Industrial and Manufacturing Engineering
  • Safety, Risk, Reliability and Quality

Fingerprint

Dive into the research topics of 'Exploring Multi-Fidelity Networks and Adapting their Architecture: A Paradigm for Enhanced Learning and Efficiency'. Together they form a unique fingerprint.

Cite this