Learning elliptic partial differential equations with randomized linear algebra
share
Given input-output pairs of an elliptic partial differential equation (PDE) in three dimensions, we derive the first theoretically-rigorous scheme for learning the associated Green's function G. By exploiting the hierarchical low-rank structure of G, we show that one can construct an approximant to G that converges almost surely and achieves an expected relative error of ϵ using at most 𝒪(ϵ^-6log^4(1/ϵ)/Γ_ϵ) input-output training pairs, for any 0<ϵ<1. The quantity 0<Γ_ϵ≤ 1 characterizes the quality of the training dataset. Along the way, we extend the randomized singular value decomposition algorithm for learning matrices to Hilbert–Schmidt operators and characterize the quality of covariance kernels for PDE learning.
READ FULL TEXT
