Randomized neural networks require a sampling domain sized to target smoothness for optimal approximation, and an adaptive PIRaNN method with partition-of-unity refinement solves PDEs with limited local regularity.
Approximation by Combinations of ReLU and Squared ReLU Ridge Functions with $ \ell^1 $ and $ \ell^0 $ Controls
2 Pith papers cite this work. Polarity classification is still indexing.
abstract
We establish $ L^{\infty} $ and $ L^2 $ error bounds for functions of many variables that are approximated by linear combinations of ReLU (rectified linear unit) and squared ReLU ridge functions with $ \ell^1 $ and $ \ell^0 $ controls on their inner and outer parameters. With the squared ReLU ridge function, we show that the $ L^2 $ approximation error is inversely proportional to the inner layer $ \ell^0 $ sparsity and it need only be sublinear in the outer layer $ \ell^0 $ sparsity. Our constructions are obtained using a variant of the Jones-Barron probabilistic method, which can be interpreted as either stratified sampling with proportionate allocation or two-stage cluster sampling. We also provide companion error lower bounds that reveal near optimality of our constructions. Despite the sparsity assumptions, we showcase the richness and flexibility of these ridge combinations by defining a large family of functions, in terms of certain spectral conditions, that are particularly well approximated by them.
verdicts
UNVERDICTED 2representative citing papers
The paper defines a distance between measures, gives an explicit recuperation operator, and proves that the resulting approximation error bounds are optimal for measures of finite total variation.
citing papers explorer
-
Adaptive Randomized Neural Networks with Locally Activation Function: Theory and Algorithm for Solving PDEs
Randomized neural networks require a sampling domain sized to target smoothness for optimal approximation, and an adaptive PIRaNN method with partition-of-unity refinement solves PDEs with limited local regularity.
-
Super-resolution meets machine learning: approximation of measures
The paper defines a distance between measures, gives an explicit recuperation operator, and proves that the resulting approximation error bounds are optimal for measures of finite total variation.