Tensor Sketch: Fast and Scalable Polynomial Kernel Approximation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pham, Ninh, Pagh, Rasmus
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910950018252800
author Pham, Ninh
Pagh, Rasmus
author_facet Pham, Ninh
Pagh, Rasmus
contents Approximation of non-linear kernels using random feature maps has become a powerful technique for scaling kernel methods to large datasets. We propose $\textit{Tensor Sketch}$, an efficient random feature map for approximating polynomial kernels. Given $n$ training samples in $\mathbb{R}^d$ Tensor Sketch computes low-dimensional embeddings in $\mathbb{R}^D$ in time $\mathcal{O}\left( n(d+D \log{D}) \right)$ making it well-suited for high-dimensional and large-scale settings. We provide theoretical guarantees on the approximation error, ensuring the fidelity of the resulting kernel function estimates. We also discuss extensions and highlight applications where Tensor Sketch serves as a central computational tool.
format Preprint
id arxiv_https___arxiv_org_abs_2505_08146
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Tensor Sketch: Fast and Scalable Polynomial Kernel Approximation
Pham, Ninh
Pagh, Rasmus
Data Structures and Algorithms
Machine Learning
Approximation of non-linear kernels using random feature maps has become a powerful technique for scaling kernel methods to large datasets. We propose $\textit{Tensor Sketch}$, an efficient random feature map for approximating polynomial kernels. Given $n$ training samples in $\mathbb{R}^d$ Tensor Sketch computes low-dimensional embeddings in $\mathbb{R}^D$ in time $\mathcal{O}\left( n(d+D \log{D}) \right)$ making it well-suited for high-dimensional and large-scale settings. We provide theoretical guarantees on the approximation error, ensuring the fidelity of the resulting kernel function estimates. We also discuss extensions and highlight applications where Tensor Sketch serves as a central computational tool.
title Tensor Sketch: Fast and Scalable Polynomial Kernel Approximation
topic Data Structures and Algorithms
Machine Learning
url https://arxiv.org/abs/2505.08146