Patent · US Active

Systems and methods for reading and writing sparse data in a neural network accelerator

US11954025B2 · kind B2 · utility

0Cited by
3References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateMar 24, 2023
Grant dateApr 9, 2024
Priority date
Expiry dateMar 24, 2043

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N3/08
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Disclosed herein includes a system, a method, and a device for reading and writing sparse data in a neural network accelerator. A mask identifying byte positions within a data word having non-zero values in memory can be accessed. Each bit of the mask can have a first value or a second value, the first value indicating that a byte of the data word corresponds to a non-zero byte value, the second value indicating that the byte of the data word corresponds to a zero byte value. The data word can be modified to have non-zero byte values stored at an end of a first side of the data word in the memory, and any zero byte values stored in a remainder of the data word. The modified data word can be written to the memory via at least a first slice of a plurality of slices that is configured to access the first side of the data word in the memory.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.