Vocabulary for Universal Approximation: A Linguistic Perspective of Mapping Compositions

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autor principal: Cai, Yongqiang
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866929352877277184
author Cai, Yongqiang
author_facet Cai, Yongqiang
contents In recent years, deep learning-based sequence modelings, such as language models, have received much attention and success, which pushes researchers to explore the possibility of transforming non-sequential problems into a sequential form. Following this thought, deep neural networks can be represented as composite functions of a sequence of mappings, linear or nonlinear, where each composition can be viewed as a \emph{word}. However, the weights of linear mappings are undetermined and hence require an infinite number of words. In this article, we investigate the finite case and constructively prove the existence of a finite \emph{vocabulary} $V=\{ϕ_i: \mathbb{R}^d \to \mathbb{R}^d | i=1,...,n\}$ with $n=O(d^2)$ for the universal approximation. That is, for any continuous mapping $f: \mathbb{R}^d \to \mathbb{R}^d$, compact domain $Ω$ and $\varepsilon>0$, there is a sequence of mappings $ϕ_{i_1}, ..., ϕ_{i_m} \in V, m \in \mathbb{Z}_+$, such that the composition $ϕ_{i_m} \circ ... \circ ϕ_{i_1} $ approximates $f$ on $Ω$ with an error less than $\varepsilon$. Our results demonstrate an unusual approximation power of mapping compositions and motivate a novel compositional model for regular languages.
format Preprint
id arxiv_https___arxiv_org_abs_2305_12205
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Vocabulary for Universal Approximation: A Linguistic Perspective of Mapping Compositions
Cai, Yongqiang
Machine Learning
Numerical Analysis
Dynamical Systems
In recent years, deep learning-based sequence modelings, such as language models, have received much attention and success, which pushes researchers to explore the possibility of transforming non-sequential problems into a sequential form. Following this thought, deep neural networks can be represented as composite functions of a sequence of mappings, linear or nonlinear, where each composition can be viewed as a \emph{word}. However, the weights of linear mappings are undetermined and hence require an infinite number of words. In this article, we investigate the finite case and constructively prove the existence of a finite \emph{vocabulary} $V=\{ϕ_i: \mathbb{R}^d \to \mathbb{R}^d | i=1,...,n\}$ with $n=O(d^2)$ for the universal approximation. That is, for any continuous mapping $f: \mathbb{R}^d \to \mathbb{R}^d$, compact domain $Ω$ and $\varepsilon>0$, there is a sequence of mappings $ϕ_{i_1}, ..., ϕ_{i_m} \in V, m \in \mathbb{Z}_+$, such that the composition $ϕ_{i_m} \circ ... \circ ϕ_{i_1} $ approximates $f$ on $Ω$ with an error less than $\varepsilon$. Our results demonstrate an unusual approximation power of mapping compositions and motivate a novel compositional model for regular languages.
title Vocabulary for Universal Approximation: A Linguistic Perspective of Mapping Compositions
topic Machine Learning
Numerical Analysis
Dynamical Systems
url https://arxiv.org/abs/2305.12205