\u5c71\u4e0b \u5b99\u4eba<\/a><\/p>\n\n\n\nAbstract:
Scientific machine learning (SciML) refers to an emerging research field that addresses scientific problems with sparse and noisy data by incorporating domain knowledge to machine learning models. The physics-informed neural network (PINN) is a representative of SciML that can solve partial differential equations (PDEs) by defining the loss function of neural networks as the residuals from the PDEs and initial\/boundary conditions. PINNs can also be used for inversion analysis and data assimilation by combining the residuals from observational data. This flexibility enables diverse applications in scientific and engineering fields. In the former part of the talk, I review the basic concepts of PINNs. In the latter part, I introduce our work on modeling crustal deformation caused by earthquakes, as a typical example of PINN applications.<\/p>\n","protected":false},"featured_media":0,"template":"","acf":[],"_links":{"self":[{"href":"http:\/\/cinetjp-static3.nict.go.jp\/japanese\/wp-json\/wp\/v2\/event\/4263"}],"collection":[{"href":"http:\/\/cinetjp-static3.nict.go.jp\/japanese\/wp-json\/wp\/v2\/event"}],"about":[{"href":"http:\/\/cinetjp-static3.nict.go.jp\/japanese\/wp-json\/wp\/v2\/types\/event"}],"wp:attachment":[{"href":"http:\/\/cinetjp-static3.nict.go.jp\/japanese\/wp-json\/wp\/v2\/media?parent=4263"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}