Patent · US Active

Wavelet representation for accelerated deep learning

US10515303B2 · kind B2 · utility

46Cited by
9References
42Claims
0Family size

Assignee

Inventors

Key dates

Filing dateApr 15, 2018
Grant dateDec 24, 2019
Priority date
Expiry dateApr 15, 2038

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N3/084
  • 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 performs flow-based computations on wavelets of data. Each processing element has a compute element with dedicated storage and a routing element. Each router enables communication with nearest neighbors in a 2D mesh. The communication is via wavelets in accordance with a representation comprising an index specifier, a virtual channel specifier, a task specifier, a data element specifier, and an optional control/data specifier. The virtual channel specifier and the task specifier are associated with one or more instructions. The index specifier and the data element are optionally associated with operands of the one or more instructions.

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