The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.

The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.

Date of Patent:
Jul. 15, 2025

Filed:

Oct. 28, 2021
Applicant:

Microsoft Technology Licensing, Llc, Redmond, WA (US);

Inventors:

Qiang Xiao, Sunnyvale, CA (US);

Haichao Wei, Santa Clara, CA (US);

Jun Shi, Fremont, CA (US);

Huiji Gao, Sunnyvale, CA (US);

Assignee:
Attorneys:
Primary Examiner:
Int. Cl.
CPC ...
G06N 20/20 (2019.01); G06F 18/21 (2023.01); G06F 18/214 (2023.01); G06N 20/10 (2019.01); G06Q 10/1053 (2023.01);
U.S. Cl.
CPC ...
G06N 20/10 (2019.01); G06F 18/214 (2023.01); G06F 18/217 (2023.01); G06N 20/20 (2019.01); G06Q 10/1053 (2013.01);
Abstract

In an example, a particular type of deep learning model is used in the global model of the GDMix model: a Factorization Machine. A Factorization Machine combines a Support Vector Machine (SVM) and Matrix Factorizations. It has the advantage of modeling data with huge sparsity well, while maintaining a linear time complexity. A modification may be further made to the Factorization Machine by introducing Lnorm reduction. This acts to divide calculations made by the Factorization Machine into a portion that can be precomputed and a portion that cannot be precomputed. The portion that can be precomputed is then precomputed in an offline manner. As such, when the model is operated in an online manner, the Factorization Machine only needs to compute the portion that cannot be precomputed, reducing the number of operations that need to performed at runtime and greatly improving processing speed over prior machine learned models.


Find Patent Forward Citations

Loading…