Distilling experience into a physically interpretable recommender system for computational model selection
Abstract Model selection is a chronic issue in computational science. The conventional approach relies heavily on human experience. However, gaining experience takes years and is severely inefficient. To address this issue, we distill human experience into a recommender system. A trained recommender...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2023-02-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-023-27426-5 |
| _version_ | 1811165933960429568 |
|---|---|
| author | Xinyi Huang Thomas Chyczewski Zhenhua Xia Robert Kunz Xiang Yang |
| author_facet | Xinyi Huang Thomas Chyczewski Zhenhua Xia Robert Kunz Xiang Yang |
| author_sort | Xinyi Huang |
| collection | DOAJ |
| description | Abstract Model selection is a chronic issue in computational science. The conventional approach relies heavily on human experience. However, gaining experience takes years and is severely inefficient. To address this issue, we distill human experience into a recommender system. A trained recommender system tells whether a computational model does well or poorly in handling a physical process. It also tells if a physical process is important for a quantity of interest. By accumulating this knowledge, the system is able to make recommendations about computational models. We showcase the power of the system by considering Reynolds-averaged-Navier–Stokes (RANS) model selection in the field of computational fluid dynamics (CFD). Since turbulence is stochastic, there is no universal RANS model, and RANS model selection has always been an issue. A working model recommending system saves fluid engineers years and allows junior CFD practitioners to make sensible model choices like senior ones. |
| first_indexed | 2024-04-10T15:44:17Z |
| format | Article |
| id | doaj.art-80f438b306344053af6898231a9eecfb |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-04-10T15:44:17Z |
| publishDate | 2023-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-80f438b306344053af6898231a9eecfb2023-02-12T12:11:19ZengNature PortfolioScientific Reports2045-23222023-02-011311910.1038/s41598-023-27426-5Distilling experience into a physically interpretable recommender system for computational model selectionXinyi Huang0Thomas Chyczewski1Zhenhua Xia2Robert Kunz3Xiang Yang4Department of Mechanical Engineering, Pennsylvania State UniversityDepartment of Mechanical Engineering, Pennsylvania State UniversityDepartment of Engineering Mechanics, Zhejiang UniversityDepartment of Mechanical Engineering, Pennsylvania State UniversityDepartment of Mechanical Engineering, Pennsylvania State UniversityAbstract Model selection is a chronic issue in computational science. The conventional approach relies heavily on human experience. However, gaining experience takes years and is severely inefficient. To address this issue, we distill human experience into a recommender system. A trained recommender system tells whether a computational model does well or poorly in handling a physical process. It also tells if a physical process is important for a quantity of interest. By accumulating this knowledge, the system is able to make recommendations about computational models. We showcase the power of the system by considering Reynolds-averaged-Navier–Stokes (RANS) model selection in the field of computational fluid dynamics (CFD). Since turbulence is stochastic, there is no universal RANS model, and RANS model selection has always been an issue. A working model recommending system saves fluid engineers years and allows junior CFD practitioners to make sensible model choices like senior ones.https://doi.org/10.1038/s41598-023-27426-5 |
| spellingShingle | Xinyi Huang Thomas Chyczewski Zhenhua Xia Robert Kunz Xiang Yang Distilling experience into a physically interpretable recommender system for computational model selection Scientific Reports |
| title | Distilling experience into a physically interpretable recommender system for computational model selection |
| title_full | Distilling experience into a physically interpretable recommender system for computational model selection |
| title_fullStr | Distilling experience into a physically interpretable recommender system for computational model selection |
| title_full_unstemmed | Distilling experience into a physically interpretable recommender system for computational model selection |
| title_short | Distilling experience into a physically interpretable recommender system for computational model selection |
| title_sort | distilling experience into a physically interpretable recommender system for computational model selection |
| url | https://doi.org/10.1038/s41598-023-27426-5 |
| work_keys_str_mv | AT xinyihuang distillingexperienceintoaphysicallyinterpretablerecommendersystemforcomputationalmodelselection AT thomaschyczewski distillingexperienceintoaphysicallyinterpretablerecommendersystemforcomputationalmodelselection AT zhenhuaxia distillingexperienceintoaphysicallyinterpretablerecommendersystemforcomputationalmodelselection AT robertkunz distillingexperienceintoaphysicallyinterpretablerecommendersystemforcomputationalmodelselection AT xiangyang distillingexperienceintoaphysicallyinterpretablerecommendersystemforcomputationalmodelselection |