Fixed-architecture networks of width O(D) and depth O(r) approximate Hölder functions with parameter magnitude log P = O(ε^{-2D/(r+γ)} log(1/ε)) via CRT encoding.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
ParaRNN decouples RNN dynamics into interpretable additive components, enabling parallelization and nonparametric regression bounds while matching vanilla RNN performance on sequential tasks.
citing papers explorer
-
On Explicit Super-Expressive Approximation for Neural Networks
Fixed-architecture networks of width O(D) and depth O(r) approximate Hölder functions with parameter magnitude log P = O(ε^{-2D/(r+γ)} log(1/ε)) via CRT encoding.
-
ParaRNN: An Interpretable and Parallelizable Recurrent Neural Network for Time-Dependent Data
ParaRNN decouples RNN dynamics into interpretable additive components, enabling parallelization and nonparametric regression bounds while matching vanilla RNN performance on sequential tasks.