Collocation approximation by deep neural ReLU networks for parametric elliptic PDEs with lognormal inputs
arXiv:2111.05504
Abstract
We obtained convergence rates of the collocation approximation by deep ReLU neural networks of solutions to elliptic PDEs with lognormal inputs, parametrized by from the non-compact set . The approximation error is measured in the norm of the Bochner space , where is the infinite tensor product standard Gaussian probability measure on and is the energy space. We also obtained similar results for the case when the lognormal inputs are parametrized on with very large dimension , and the approximation error is measured in the -weighted uniform norm of the Bochner space , where is the density function of the standard Gaussian probability measure on .
References in corpus (7)
- DGM: A deep learning algorithm for solving partial differential equations
- On the Number of Linear Regions of Deep Neural Networks
- Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification
- Optimal approximation of continuous functions by very deep ReLU networks
- Representation Benefits of Deep Feedforward Networks
- Adaptivity of deep ReLU network for learning in Besov and mixed smooth Besov spaces: optimal rate and curse of dimensionality
- Nonlinear Approximation and (Deep) ReLU Networks