Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387667062> ?p ?o ?g. }
- W4387667062 endingPage "112549" @default.
- W4387667062 startingPage "112549" @default.
- W4387667062 abstract "We present a general kernel-based framework for learning operators between Banach spaces along with a priori error analysis and comprehensive numerical comparisons with popular neural net (NN) approaches such as Deep Operator Networks (DeepONet) [46] and Fourier Neural Operator (FNO) [45]. We consider the setting where the input/output spaces of target operator G†:U→V are reproducing kernel Hilbert spaces (RKHS), the data comes in the form of partial observations ϕ(ui),φ(vi) of input/output functions vi=G†(ui) (i=1,…,N), and the measurement operators ϕ:U→Rn and φ:V→Rm are linear. Writing ψ:Rn→U and χ:Rm→V for the optimal recovery maps associated with ϕ and φ, we approximate G† with G¯=χ∘f¯∘ϕ where f¯ is an optimal recovery approximation of f†:=φ∘G†∘ψ:Rn→Rm. We show that, even when using vanilla kernels (e.g., linear or Matérn), our approach is competitive in terms of cost-accuracy trade-off and either matches or beats the performance of NN methods on a majority of benchmarks. Additionally, our framework offers several advantages inherited from kernel methods: simplicity, interpretability, convergence guarantees, a priori error estimates, and Bayesian uncertainty quantification. As such, it can serve as a natural benchmark for operator learning." @default.
- W4387667062 created "2023-10-17" @default.
- W4387667062 creator A5012461935 @default.
- W4387667062 creator A5017359648 @default.
- W4387667062 creator A5034000581 @default.
- W4387667062 creator A5057629021 @default.
- W4387667062 date "2023-10-01" @default.
- W4387667062 modified "2023-10-17" @default.
- W4387667062 title "Kernel Methods are Competitive for Operator Learning" @default.
- W4387667062 cites W1540155273 @default.
- W4387667062 cites W1856502440 @default.
- W4387667062 cites W1972576201 @default.
- W4387667062 cites W1973243996 @default.
- W4387667062 cites W1973333099 @default.
- W4387667062 cites W1980404981 @default.
- W4387667062 cites W1982421072 @default.
- W4387667062 cites W1991788899 @default.
- W4387667062 cites W1998576042 @default.
- W4387667062 cites W2005606401 @default.
- W4387667062 cites W2007650703 @default.
- W4387667062 cites W2018159038 @default.
- W4387667062 cites W2021012767 @default.
- W4387667062 cites W2032768921 @default.
- W4387667062 cites W2041099140 @default.
- W4387667062 cites W2042938036 @default.
- W4387667062 cites W2056227133 @default.
- W4387667062 cites W2071519111 @default.
- W4387667062 cites W2072205359 @default.
- W4387667062 cites W2080556931 @default.
- W4387667062 cites W2082261407 @default.
- W4387667062 cites W2083845086 @default.
- W4387667062 cites W2092398714 @default.
- W4387667062 cites W2117188967 @default.
- W4387667062 cites W2133041109 @default.
- W4387667062 cites W2139923370 @default.
- W4387667062 cites W2157890204 @default.
- W4387667062 cites W2160812957 @default.
- W4387667062 cites W2161024376 @default.
- W4387667062 cites W2216523920 @default.
- W4387667062 cites W2465076683 @default.
- W4387667062 cites W2621213434 @default.
- W4387667062 cites W2766298346 @default.
- W4387667062 cites W2784733489 @default.
- W4387667062 cites W2885076212 @default.
- W4387667062 cites W2887569307 @default.
- W4387667062 cites W2896618593 @default.
- W4387667062 cites W2963634130 @default.
- W4387667062 cites W2963798430 @default.
- W4387667062 cites W2979712029 @default.
- W4387667062 cites W2981014499 @default.
- W4387667062 cites W2989600087 @default.
- W4387667062 cites W3095387974 @default.
- W4387667062 cites W3098175809 @default.
- W4387667062 cites W3127355902 @default.
- W4387667062 cites W3128400532 @default.
- W4387667062 cites W3137474564 @default.
- W4387667062 cites W3155546189 @default.
- W4387667062 cites W3156927273 @default.
- W4387667062 cites W3164726984 @default.
- W4387667062 cites W3168833755 @default.
- W4387667062 cites W3178968719 @default.
- W4387667062 cites W3179393474 @default.
- W4387667062 cites W3196020781 @default.
- W4387667062 cites W3198272795 @default.
- W4387667062 cites W3198653666 @default.
- W4387667062 cites W3201666041 @default.
- W4387667062 cites W3202809124 @default.
- W4387667062 cites W3205503552 @default.
- W4387667062 cites W3209120121 @default.
- W4387667062 cites W3217167116 @default.
- W4387667062 cites W3217544405 @default.
- W4387667062 cites W4206212643 @default.
- W4387667062 cites W4211049957 @default.
- W4387667062 cites W4221019179 @default.
- W4387667062 cites W4225401869 @default.
- W4387667062 cites W4283460575 @default.
- W4387667062 cites W4296754083 @default.
- W4387667062 cites W4309879846 @default.
- W4387667062 doi "https://doi.org/10.1016/j.jcp.2023.112549" @default.
- W4387667062 hasPublicationYear "2023" @default.
- W4387667062 type Work @default.
- W4387667062 citedByCount "0" @default.
- W4387667062 crossrefType "journal-article" @default.
- W4387667062 hasAuthorship W4387667062A5012461935 @default.
- W4387667062 hasAuthorship W4387667062A5017359648 @default.
- W4387667062 hasAuthorship W4387667062A5034000581 @default.
- W4387667062 hasAuthorship W4387667062A5057629021 @default.
- W4387667062 hasConcept C104317684 @default.
- W4387667062 hasConcept C111472728 @default.
- W4387667062 hasConcept C11413529 @default.
- W4387667062 hasConcept C118615104 @default.
- W4387667062 hasConcept C126255220 @default.
- W4387667062 hasConcept C138885662 @default.
- W4387667062 hasConcept C154945302 @default.
- W4387667062 hasConcept C158448853 @default.
- W4387667062 hasConcept C17020691 @default.
- W4387667062 hasConcept C185592680 @default.
- W4387667062 hasConcept C202444582 @default.