US 11,657,286 B2
Structure learning in convolutional neural networks
Andrew Rabinovich, San Francisco, CA (US); Vijay Badrinarayanan, Mountain View, CA (US); Daniel DeTone, San Francisco, CA (US); Srivignesh Rajendran, Mountain View, CA (US); Douglas Bertram Lee, Redwood City, CA (US); and Tomasz Malisiewicz, Mountain View, CA (US)
Assigned to Magic Leap, Inc., Plantation, FL (US)
Filed by MAGIC LEAP, INC., Plantation, FL (US)
Filed on Feb. 23, 2021, as Appl. No. 17/183,021.
Application 17/183,021 is a continuation of application No. 16/366,047, filed on Mar. 27, 2019, granted, now 10,963,758.
Application 16/366,047 is a continuation of application No. 15/457,990, filed on Mar. 13, 2017, granted, now 10,255,529, issued on Apr. 9, 2019.
Claims priority of provisional application 62/307,071, filed on Mar. 11, 2016.
Prior Publication US 2021/0182636 A1, Jun. 17, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 9/00 (2022.01); G06V 30/194 (2022.01); G06N 3/082 (2023.01); G06V 10/44 (2022.01); G06F 18/24 (2023.01); G06F 18/2413 (2023.01); G06N 3/045 (2023.01)
CPC G06V 30/194 (2022.01) [G06F 18/24 (2023.01); G06F 18/24137 (2023.01); G06N 3/045 (2023.01); G06N 3/082 (2013.01); G06V 10/454 (2022.01)] 19 Claims
OG exemplary drawing
 
1. A method implemented with a processor, comprising:
creating a neural network comprising a plurality of layers;
identifying an extraneous layer from the plurality of layers;
removing the extraneous layer from the neural network; and
identifying and removing other extraneous layers from the plurality of layers until no other extraneous layer is identified,
wherein the neural network undergoes both vertical splitting and horizontal splitting.