EXPLOITING INPUT DATA SPARSITY IN NEURAL NETWORK COMPUTE UNITS

Patent №

US 11,106,606

Granted

2021-08-31

Filed 2019

Owner

GOOGLE INC.

AI components

2

ml · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

16514562

A computer-implemented method includes receiving, by a computing device, input activations and determining, by a controller of the computing device, whether each of the input activations has either a zero value or a non-zero value. The method further includes storing, in a memory bank of the computing device, at least one of the input activations. Storing the at least one input activation includes generating an index comprising one or more memory address locations that have input activation values that are non-zero values. The method still further includes providing, by the controller and from the memory bank, at least one input activation onto a data bus that is accessible by one or more units of a computational array. The activations are provided, at least in part, from a memory address location associated with the index.

Machine learningAI hardwareG06F 13/1668G06F 9/3001G06F 9/38G06F 9/3824G06F 15/76G06F 17/16G06N 3/045G06N 3/0464+7 more

AI classification

AI hardware1.00
Machine learning1.00
Vision0.01
Evolutionary computation0.00
Natural language0.00
Knowledge representation0.00
Speech0.00
Planning0.00

Ownership

GOOGLE INC.

assignment · 497980404

Assignors

WOO, DONG HYUK, NARAYANASWAMI, RAVI

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC