Gradient Testing and Estimation by Comparisons

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Tao, Xiwen, Zhang, Chenyi, Wang, Helin, Zhang, Yexin, Li, Tongyang
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908840826503168
author Tao, Xiwen
Zhang, Chenyi
Wang, Helin
Zhang, Yexin
Li, Tongyang
author_facet Tao, Xiwen
Zhang, Chenyi
Wang, Helin
Zhang, Yexin
Li, Tongyang
contents We study gradient testing and gradient estimation of smooth functions using only a comparison oracle that, given two points, indicates which one has the larger function value. For any smooth $f\colon\mathbb R^n\to\mathbb R$, $\mathbf{x}\in\mathbb R^n$, and $\varepsilon>0$, we design a gradient testing algorithm that determines whether the normalized gradient $\nabla f(\mathbf{x})/\|\nabla f(\mathbf{x})\|$ is $\varepsilon$-close or $2\varepsilon$-far from a given unit vector $\mathbf{v}$ using $O(1)$ queries, as well as a gradient estimation algorithm that outputs an $\varepsilon$-estimate of $\nabla f(\mathbf{x})/\|\nabla f(\mathbf{x})\|$ using $O(n\log(1/\varepsilon))$ queries which we prove to be optimal. Furthermore, we study gradient estimation in the quantum comparison oracle model where queries can be made in superpositions, and develop a quantum algorithm using $O(\log (n/\varepsilon))$ queries.
format Preprint
id arxiv_https___arxiv_org_abs_2405_11454
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Gradient Testing and Estimation by Comparisons
Tao, Xiwen
Zhang, Chenyi
Wang, Helin
Zhang, Yexin
Li, Tongyang
Machine Learning
Data Structures and Algorithms
Optimization and Control
We study gradient testing and gradient estimation of smooth functions using only a comparison oracle that, given two points, indicates which one has the larger function value. For any smooth $f\colon\mathbb R^n\to\mathbb R$, $\mathbf{x}\in\mathbb R^n$, and $\varepsilon>0$, we design a gradient testing algorithm that determines whether the normalized gradient $\nabla f(\mathbf{x})/\|\nabla f(\mathbf{x})\|$ is $\varepsilon$-close or $2\varepsilon$-far from a given unit vector $\mathbf{v}$ using $O(1)$ queries, as well as a gradient estimation algorithm that outputs an $\varepsilon$-estimate of $\nabla f(\mathbf{x})/\|\nabla f(\mathbf{x})\|$ using $O(n\log(1/\varepsilon))$ queries which we prove to be optimal. Furthermore, we study gradient estimation in the quantum comparison oracle model where queries can be made in superpositions, and develop a quantum algorithm using $O(\log (n/\varepsilon))$ queries.
title Gradient Testing and Estimation by Comparisons
topic Machine Learning
Data Structures and Algorithms
Optimization and Control
url https://arxiv.org/abs/2405.11454