Deep Learning for Generalized Multiobjective Optimization of Metamaterials

R. P. Jenkins, P. J. O'Connor, S. D. Campbell, P. L. Werner, D. H. Werner

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

Abstract

The inverse-design of metamaterials often requires full-wave evaluation of an enormous number of different candidate designs, which can be extremely time consuming and, in many cases of practical interest, even intractable. By introducing deep learning at an early stage in the design process, a generalized network can be paired with multiobjective optimization to rapidly solve a variety of complex electromagnetic metamaterial design problems.

Original languageEnglish (US)
Title of host publication2020 14th International Congress on Artificial Materials for Novel Wave Phenomena, Metamaterials 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages261-263
Number of pages3
ISBN (Electronic)9781728161044
DOIs
StatePublished - Sep 27 2020
Event14th International Congress on Artificial Materials for Novel Wave Phenomena, Metamaterials 2020 - New York City, United States
Duration: Sep 27 2020Oct 3 2020

Publication series

Name2020 14th International Congress on Artificial Materials for Novel Wave Phenomena, Metamaterials 2020

Conference

Conference14th International Congress on Artificial Materials for Novel Wave Phenomena, Metamaterials 2020
Country/TerritoryUnited States
CityNew York City
Period9/27/2010/3/20

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Acoustics and Ultrasonics
  • Atomic and Molecular Physics, and Optics
  • Radiation

Fingerprint

Dive into the research topics of 'Deep Learning for Generalized Multiobjective Optimization of Metamaterials'. Together they form a unique fingerprint.

Cite this