← Search

Dmytro Perekrestenko

2 accepted papers

2020

Constructive Universal High-Dimensional Distribution Generation through Deep ReLU Networks

ICML 2020poster

We present an explicit deep neural network construction that transforms uniformly distributed one-dimensional noise into an arbitrarily close approximation of any two-dimensional Lipschitz-continuous target distribution. The key ingredient of our design is a generalization of the "space-filling" pro…

Cited by 21SourcePDFScholar