Work

Code supplement: Second-order cumulant corrections for ReLU random-feature kernels under non-Gaussian weights

Year: 2026

Type: software

Abstract: Code, frozen specifications, result files with hashes and the Lean project behind the article "Second-order cumulant corrections for ReLU random-feature kernels under non-Gaussian weights: accuracy, cost and limits" (S. Zabolotnii, V. Khotunov, P. Mulesa). Every number, table and figure of the article is regenerated from the stored results by paper/scripts/make_tex.py; see README.md.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.23756873

Authors: Zabolotnii, Serhii, Khotunov, Vladyslav, Mulesa, Pavlo

Institutions: Cherkasy State Business College, Uzhhorod National University

Language: English

Topic: Stochastic Gradient Optimization Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7224388917

Cites:

Cited by:

Related to: 10