Patent · US Active

Data operations and finite state machine for machine learning via bypass of computational tasks based on frequently-used data values

US11269643B2 · kind B2 · utility

2Cited by

10References

16Claims

0Family size

Assignee

Intel Corporation · US

Inventors

Liwei Ma · Beijing, CN
Nadathur Rajagopalan Satish · Santa Clara, US
Jeremy Bottleson · Rancho Cordova, US
Farshad Akhbari · Chandler, US
Eriko Nurvitadhi · Beaverton, US
Abhishek R. Appu · Cameron Park, US
Altug Koker · Cameron Park, US
Kamal Sinha · Rancho Cordova, US
Joydeep Ray · Folsom, US
Balaji Vembu · Folsom, US
Vasanth Ranganathan · Cameron Park, US
Sanjeev Jahagirdar · Folsom, US

Key dates

Filing date	Apr 9, 2017
Grant date	Mar 8, 2022
Priority date	—
Expiry date	Jan 10, 2040

Classification

Technology area (CPC G)Physics
CPC primaryG06T1/20
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A mechanism is described for facilitating fast data operations and for facilitating a finite state machine for machine learning at autonomous machines. A method of embodiments, as described herein, includes detecting input data to be used in computational tasks by a computation component of a processor including a graphics processor. The method may further include determining one or more frequently-used data values (FDVs) from the data, and pushing the one or more frequent data values to bypass the computational tasks.

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