US 11,853,865 B2
Prefetching weights for use in a neural network processor
Jonathan Ross, Mountain View, CA (US)
Assigned to Google LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Dec. 28, 2020, as Appl. No. 17/134,936.
Application 17/134,936 is a continuation of application No. 16/826,466, filed on Mar. 23, 2020, granted, now 10,878,316.
Application 16/826,466 is a continuation of application No. 16/053,305, filed on Aug. 2, 2018, granted, now 11,281,966.
Application 16/053,305 is a continuation of application No. 14/844,670, filed on Sep. 3, 2015, granted, now 10,049,322, issued on Aug. 14, 2018.
Claims priority of provisional application 62/164,981, filed on May 21, 2015.
Prior Publication US 2021/0192328 A1, Jun. 24, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/063 (2023.01); G06F 15/80 (2006.01)
CPC G06N 3/063 (2013.01) [G06F 15/8046 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system for performing neural network computations for a neural network having a plurality of neural network layers, the system comprising:
a matrix computation unit comprising an array of cells wherein each cell includes respective circuitry configured to:
obtain a respective weight input for a neural network layer of the plurality of neural network layers;
receive a control signal from a sequencer; and
determine, based on the control signal, whether to reuse the respective weight input for another cell at a subsequent clock cycle.