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:
Mar. 24, 2026

Filed:

Dec. 31, 2018
Applicant:

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

Inventors:

Bita Darvish Rouhani, Bellevue, WA (US);

Eric S. Chung, Woodinville, WA (US);

Daniel Lo, Bothell, WA (US);

Douglas C. Burger, Bellevue, WA (US);

Assignee:
Attorney:
Primary Examiner:
Int. Cl.
CPC ...
G06N 3/063 (2023.01); G06F 9/30 (2018.01); G06F 18/21 (2023.01); G06N 3/047 (2023.01); G06N 3/084 (2023.01);
U.S. Cl.
CPC ...
G06N 3/063 (2013.01); G06F 9/30025 (2013.01); G06F 18/217 (2023.01); G06N 3/047 (2023.01); G06N 3/084 (2013.01);
Abstract

Apparatus and methods for training neural networks based on a performance metric, including adjusting numerical precision and topology as training progresses are disclosed. In some examples, block floating-point formats having relatively lower accuracy are used during early stages of training. Accuracy of the floating-point format can be increased as training progresses based on a determined performance metric. In some examples, values for the neural network are transformed to normal precision floating-point formats. The performance metric can be determined based on entropy of values for the neural network, accuracy of the neural network, or by other suitable techniques. Accelerator hardware can be used to implement certain implementations, including hardware having direct support for block floating-point formats.


Find Patent Forward Citations

Loading…