Patent · US Active

Method of enabling sparse neural networks on memresistive accelerators

US11816563B2 · kind B2 · utility

0Cited by
1References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateMay 10, 2019
Grant dateNov 14, 2023
Priority date
Expiry dateMar 26, 2042

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N3/0495
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A method of storing a sparse weight matrix for a trained artificial neural network in a circuit including a series of clusters. The method includes partitioning the sparse weight matrix into at least one first sub-block and at least one second sub-block. The first sub-block includes only zero-value weights and the second sub-block includes non-zero value weights. The method also includes assigning the non-zero value weights in the at least one second sub-block to at least one cluster of the series of clusters of the circuit. The circuit is configured to perform matrix-vector-multiplication (MVM) between the non-zero value weights of the at least one second sub-block and an input vector during an inference process utilizing the artificial neural network. The sub-blocks containing all zero elements are power gated, thereby reducing overall energy consumption for inference.

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