Block floating point computations using shared exponents
US10579334B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | May 8, 2018 |
| Grant date | Mar 3, 2020 |
| Priority date | — |
| Expiry date | May 19, 2038 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F17/16
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A system for block floating point computation in a neural network receives a plurality of floating point numbers. An exponent value for an exponent portion of each floating point number of the plurality of floating point numbers is identified and mantissa portions of the floating point numbers are grouped. A shared exponent value of the grouped mantissa portions is selected according to the identified exponent values and then removed from the grouped mantissa portions to define multi-tiered shared exponent block floating point numbers. One or more dot product operations are performed on the grouped mantissa portions of the multi-tiered shared exponent block floating point numbers to obtain individual results. The individual results are shifted to generate a final dot product value, which is used to implement the neural network. The shared exponent block floating point computations reduce processing time with less reduction in system accuracy.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.