	US 11,934,791 B2
	On-device projection neural networks for natural language understanding
Sujith Ravi, Santa Clara, CA (US); and Zornitsa Kozareva, Santa Clara, CA (US)
Assigned to GOOGLE LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Aug. 1, 2022, as Appl. No. 17/878,631.
Application 17/878,631 is a continuation of application No. 17/141,473, filed on Jan. 5, 2021, granted, now 11,423,233.
Application 17/141,473 is a continuation of application No. 16/135,545, filed on Sep. 19, 2018, granted, now 10,885,277, issued on Jan. 5, 2021.
Claims priority of provisional application 62/714,028, filed on Aug. 2, 2018.
Prior Publication US 2023/0048218 A1, Feb. 16, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 40/30 (2020.01); G06F 40/253 (2020.01); G06N 3/04 (2023.01); G06N 3/084 (2023.01)

CPC G06F 40/30 (2020.01) [G06F 40/253 (2020.01); G06N 3/04 (2013.01); G06N 3/084 (2013.01)]

20 Claims

1. A computing device configured to perform text or speech classification, the computing device comprising:

a projection neural network configured to receive a projection network input and to generate a projection network output from the projection network input;

wherein the projection network input comprises data descriptive of one or more units of text or speech;

wherein the projection network output comprises one or more classification outputs for the one or more units of text or speech; and

wherein the projection neural network comprises:

a sequence of one or more projection layers, wherein each projection layer has a plurality of projection layer parameters, wherein each projection layer is configured to:

receive a layer input;

apply a plurality of projection layer functions to the layer input, each projection layer function generating a respective projection function output that projects the layer input to a different space, and

generate a layer output by applying the projection layer parameters for the projection layer to the projection function outputs.