ReLU Network with Width $d+\mathcalO(1)$ Can Achieve Optimal Approximation Rate

Authors: Chenghao Liu, Minghua Chen

ICML 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper, we address this gap by proving that Re LU networks with width d + 1 can achieve the optimal approximation rate for continuous functions over the domain [0, 1]d under Lp norm for p [1, ). We further show that for the uniform norm, a width of d + 11 is sufficient. We also extend the results to narrow feed-forward networks with various activations, confirming their capability to approximate at the optimal rate. This work adds to the understanding of universal approximation of narrow networks.
Researcher Affiliation Academia 1School of Data Science, City University of Hong Kong. Correspondence to: Minghua Chen <minghua.chen@cityu.edu.hk>.
Pseudocode No The paper focuses on theoretical proofs and constructions but does not include any pseudocode or clearly labeled algorithm blocks.
Open Source Code No The paper does not provide any information or links regarding the availability of open-source code for the described methodology.
Open Datasets No As a theoretical paper, it does not involve training on datasets or mention any public datasets, links, or citations for such use.
Dataset Splits No The paper is theoretical and does not conduct experiments involving validation dataset splits.
Hardware Specification No As a theoretical paper, no hardware specifications for running experiments are mentioned.
Software Dependencies No As a theoretical paper, no specific software dependencies with version numbers are mentioned.
Experiment Setup No As a theoretical paper, no experimental setup details like hyperparameter values or training configurations are provided.