Search NASA⌕ Search
Home/Archive/DOE OSTI

DOE OSTI · 3391495

HPC-Driven Modeling with ML-Based Surrogates for Magnon-Photon Dynamics in Hybrid Quantum System

Song, Jialin [Simon Fraser University, Burnaby, BC (Canada)]·Tang, Yingheng [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)]·Ren, Pu [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)]·Takayoshi, Shintaro [University of Wisconsin, Madison, WI (United States)]·Sawant, Saurabh [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)]·Zhu, Yujie [University of Wisconsin, Madison, WI (United States)]·Hu, Jia-Mian [University of Wisconsin, Madison, WI (United States)]·Nonaka, Andy [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); University of California, Berkeley, CA (United States). International Computer Science Institute (ICSI)]·Mahoney, Michael W. [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); University of California, Berkeley, CA (United States)]·Erichson, Benjamin [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); University of California, Berkeley, CA (United States). International Computer Science Institute (ICSI)]·Yao, Zhi Jackie [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)]
View original source ↗Read source document ↗Export metadata ↓

Abstract

Here, we introduce a hybrid computational framework that merges HPC-based numerical solvers with physics-informed ML surrogates for efficient modeling of magnon-photon interactions. By running short-duration, high-fidelity Maxwell-LLG simulations and feeding their results into an ML model, we substantially cut simulation time while achieving accurate predictions across larger spatiotemporal domains.

Explore related subjects

AccuracyComputational efficiencyComputational electromagneticsComputational modelingNumerical modelsPredictive modelsQuantum systemSpatiotemporal phenomena

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Song, Jialin [Simon Fraser University, Burnaby, BC (Canada)], Tang, Yingheng [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)], Ren, Pu [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)], Takayoshi, Shintaro [University of Wisconsin, Madison, WI (United States)], Sawant, Saurabh [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)], Zhu, Yujie [University of Wisconsin, Madison, WI (United States)], Hu, Jia-Mian [University of Wisconsin, Madison, WI (United States)], Nonaka, Andy [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); University of California, Berkeley, CA (United States). International Computer Science Institute (ICSI)], Mahoney, Michael W. [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); University of California, Berkeley, CA (United States)], Erichson, Benjamin [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); University of California, Berkeley, CA (United States). International Computer Science Institute (ICSI)], Yao, Zhi Jackie [Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States)]. 2025-05-18. HPC-Driven Modeling with ML-Based Surrogates for Magnon-Photon Dynamics in Hybrid Quantum System. https://doi.org/10.23919/aces66556.2025.11052493

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

NASA NTRS1975

CARETS: A prototype regional environmental information system. Volume 6: Cost, accuracy and consistency comparisons of land use maps made from high-altitude aircraft photography and ERTS imagery

Accuracy analyses for the land use maps of the Central Atlantic Regional Ecological Test Site were performed for a 1-percent sample of the area. Researchers compared Level II land use maps produced at three scales, 1:24,000, 1:100,000, and 1:250,000 from high-altitude photography, with each other and with point data obtained in the field. They employed the same procedures to determine the accuracy of the Level I land use maps produced at 1:250,000 from high-altitude photography and color composite ERTS imagery. The accuracy of the Level II maps was 84.9 percent at 1:24,000, 77.4 percent at 1:100,000, and 73.0 percent at 1:250,000. The accuracy of the Level I 1:250,000 maps produced from high-altitude aircraft photography was 76.5 percent and for those produced from ERTS imagery was 69.5 percent. The cost of Level II land use mapping at 1:24,000 was found to be high ($11.93 per km 2 ). The cost of mapping at 1:100,000 ($1.75) was about 2 times as expensive as mapping at 1:250,000 ($.88), and the accuracy increased by only 4.4 percent. Level I land use maps, when mapped from high-altitude photography, were about 4 times as expensive as the maps produced from ERTS imagery, although the accuracy is 7.0 percent greater. The Level I land use category that is least accurately mapped from ERTS imagery is urban and built-up land in the non-urban areas; in the urbanized areas, built-up land is more reliably map

Accuracy↗