A framework for a posteriori error certification of neural network approximations to elliptic and parabolic PDEs via Riesz representations of residuals on geometrically simpler domains.
Error bounds for approximations with deep ReLU neural networks in W s,p norms,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.NA 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A posteriori certification for neural network approximations to PDEs
A framework for a posteriori error certification of neural network approximations to elliptic and parabolic PDEs via Riesz representations of residuals on geometrically simpler domains.