TorchKM: A GPU-Oriented Library for Kernel Learning and Model Selection

Yikai Zhang; Gaoxiang Jia; Jie Ding; Boxiang Wang

doi:10.48550/arxiv.2606.06742

Back

TorchKM: A GPU-Oriented Library for Kernel Learning and Model Selection

Preprint

Open access

TorchKM: A GPU-Oriented Library for Kernel Learning and Model Selection

Yikai Zhang, Gaoxiang Jia, Jie Ding and Boxiang Wang

ArXiv.org

Cornell University

06/09/2026

DOI: 10.48550/arxiv.2606.06742

Files and links (1)

url

https://doi.org/10.48550/arxiv.2606.06742View

Preprint (Author's original) This preprint has not been evaluated by subject experts through peer review. Preprints may undergo extensive changes and/or become peer-reviewed journal articles. Open Access

Abstract

TorchKM is an open-source library for kernel machines, including support vector machines, kernel logistic regression, and kernel quantile regression, with GPU acceleration. The library features a scikit-learn-style API and is designed to exploit GPU-friendly linear algebra, accelerating the full training and model-selection pipeline through intelligent reuse of matrix operations. Benchmarks show competitive predictive performance with substantial speedups over standard baselines. The efficiency and programmable design also make TorchKM a kernel-learning component for AI-driven workflows. Code and documentation are available at https://github.com/YikaiZhang95/torchkm, and the package can be easily installed via PyPI.

Computer Science - Learning

Statistics - Machine Learning

Details

Title: Subtitle: TorchKM: A GPU-Oriented Library for Kernel Learning and Model Selection
Creators: Yikai Zhang - University of Iowa
Gaoxiang Jia
Jie Ding
Boxiang Wang - University of Iowa
Resource Type: Preprint
Publication Details: ArXiv.org
DOI: 10.48550/arxiv.2606.06742
ISSN: 2331-8422
Publisher: Cornell University; Ithaca, New York
Language: English
Date posted: 06/09/2026
Academic Unit: Statistics and Actuarial Science
Record Identifier: 9985174493602771

Metrics

1 Record Views