Patent · US Active

Compression technique for deep neural network weights

US11757469B2 · kind B2 · utility

1Cited by
1References
30Claims
0Family size

Assignee

Inventors

Key dates

Filing dateApr 1, 2021
Grant dateSep 12, 2023
Priority date
Expiry dateMay 12, 2042

Classification

  • Technology area (CPC H)Electricity
  • CPC primaryH03M7/46
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Various embodiments include methods and devices for compression and decompression of weight data sets. Some embodiments may include compressing weight data by receiving a weight data set of binary numbers representing weight values, generating a frame payload including a compressed first frame of a first subset of the weight values in the weight data set, and generating a block of compressed weight data having the frame payload. Some embodiments may include decompressing weight data by retrieving a block of compressed weight data, in which the block of compressed weight data includes a frame header associated with a frame payload, in which the frame header includes a normalization factor indicator, and in which the frame payload includes compressed weight values, and generating a first decompressed frame comprising decompressed weight values of the compressed weight values of the frame payload.

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