Patent · US Active

Numerical representation for neural networks

US11062202B2 · kind B2 · utility

8Cited by
0References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJul 17, 2019
Grant dateJul 13, 2021
Priority date
Expiry dateJul 17, 2039

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N3/09
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Techniques in advanced deep learning provide improvements in one or more of accuracy, performance, and energy efficiency. An array of processing elements comprising a portion of a neural network accelerator performs flow-based computations on wavelets of data. Each processing element has a respective compute element and a respective routing element. Each compute element has a respective floating-point unit enabled to optionally and/or selectively perform floating-point operations in accordance with a programmable exponent bias and/or various floating-point computation variations. In some circumstances, the programmable exponent bias and/or the floating-point computation variations enable neural network processing with improved accuracy, decreased training time, decreased inference latency, and/or increased energy efficiency.

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