公开(公告)号：US11653467B2

公开(公告)日：2023-05-16

申请号：US17251697

申请日：2018-12-21

Applicant: Intel Corporation

Inventor： Chris D. Lucero ,  Khine Han ,  Joshua D. Heppner ,  Christopher Rossi ,  Hadi Sharifi ,  Aniekeme Udofia ,  Abdul Bailey ,  Katherine Perkins ,  Kevin Lowell Hudson ,  Roderick E. Kronschnabel ,  Neha Purushothaman

IPC: H05K7/14 ,  G06F15/78 ,  H05K1/14

CPC classification number: H05K7/1422 ,  G06F15/7803 ,  H05K1/141

Abstract: An Internet of Things (IoT) apparatus including a plurality of boards and one or more connectors to couple IoT modules to one or more of the plurality of boards and to couple the plurality of boards to each other. The connectors include stacking connectors on both sides of at least some of the boards and at least some of the IoT modules to be coupled to the boards.

公开(公告)号：US20210307189A1

公开(公告)日：2021-09-30

申请号：US17251697

申请日：2018-12-21

Applicant: Intel Corporation

Inventor： Chris D. Lucero ,  Khine Han ,  Joshua D. Heppner ,  Christopher Rossi ,  Hadi Sharifi ,  Aniekeme Udofia ,  Abdul Bailey ,  Katherine Perkins ,  Kevin Lowell Hudson ,  Roderick E. Kronschnabel ,  Neha Purushothaman

IPC: H05K7/14 ,  G06F15/78 ,  H05K1/14

Abstract: In some examples, an Internet of Things (IoT) apparatus including a plurality of boards and one or more connectors to couple IoT modules to one or more of the plurality of boards and to couple the plurality of boards to each other. The connectors include stacking connectors on both sides of at least some of the boards and at least some of the IoT modules to be coupled to the boards. The stacking connectors allow the IoT modules and the boards to be coupled together in a manner that boards and modules cannot be inserted incorrectly.

公开(公告)号：US20240152756A1

公开(公告)日：2024-05-09

申请号：US18548805

申请日：2022-03-25

Applicant: Intel Corporation

Inventor： Barath Lakshmanan ,  Ashish B. Datta ,  Craig D. Sperry ,  David J. Austin ,  Caleb Mark McMillan ,  Neha Purushothaman ,  Rita H. Wouhaybi

IPC: G06N3/08

CPC classification number: G06N3/08

Abstract: In one embodiment, a method of training an autoencoder neural network includes determining autoencoder design parameters for the autoencoder neural network, including an input image size for an input image, a compression ratio for compression of the input image into a latent vector, and a latent vector size for the latent vector. The input image size is determined based on a resolution of training images and a size of target features to be detected. The compression ratio is determined based on entropy of the training images. The latent vector size is determined based on the compression ratio. The method further includes training the autoencoder neural network based on the autoencoder design parameters and the training dataset, and then saving the trained autoencoder neural network on a storage device.