公开(公告)号：US20240013098A1

公开(公告)日：2024-01-11

申请号：US18370524

申请日：2023-09-20

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Changzheng ZHANG ,  Xiaolong BAI ,  Dandan TU

IPC: G06N20/00 ,  G06F17/16 ,  G06N7/00

CPC classification number: G06N20/00 ,  G06F17/16 ,  G06N7/00 ,  G06N3/08

Abstract: Embodiments of the present invention disclose a data processing apparatus. The apparatus is configured to: after calculating a set of gradient information of each parameter by using a sample data subset, delete the sample data subset, read a next sample data subset, calculate another set of gradient information of each parameter by using the next sample data subset, and accumulate a plurality of sets of calculated gradient information of each parameter, to obtain an update gradient of each parameter.

公开(公告)号：US20190287022A1

公开(公告)日：2019-09-19

申请号：US16432617

申请日：2019-06-05

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Changzheng ZHANG ,  Xiaolong BAI ,  Dandan TU

IPC: G06N20/00 ,  G06F17/16

Abstract: Embodiments of the present invention disclose a data processing apparatus. The apparatus is configured to: after calculating a set of gradient information of each parameter by using a sample data subset, delete the sample data subset, read a next sample data subset, calculate another set of gradient information of each parameter by using the next sample data subset, and accumulate a plurality of sets of calculated gradient information of each parameter, to obtain an update gradient of each parameter.

公开(公告)号：US20220156931A1

公开(公告)日：2022-05-19

申请号：US17590005

申请日：2022-02-01

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Yaoxin LI ,  Changzheng ZHANG ,  Xiaoshi CHEN ,  Dandan TU

IPC: G06T7/00 ,  G06T7/11 ,  G06V10/764 ,  G06T7/70 ,  G06V10/774 ,  G06V10/80 ,  G06V10/20 ,  G16H30/40 ,  G16H50/20

Abstract: This application relates to the artificial intelligence field, and provides an image processing method, an apparatus, and a system. The image processing method includes: obtaining a plurality of image blocks by segmenting a to-be-analyzed pathological image; inputting the plurality of image blocks to a first analysis model to obtain a first analysis result, where the first analysis model classifies each of the plurality of image blocks based on a quantity or an area of suspicious lesion components; inputting at least one second-type image block in the first analysis result to a second analysis model to obtain a second analysis result, where the second analysis model analyzes a location of a suspicious lesion component of each input second-type image block; and obtaining a final analysis result of the pathological image based on the first analysis result and the second analysis result.

公开(公告)号：US20190279088A1

公开(公告)日：2019-09-12

申请号：US16424760

申请日：2019-05-29

Applicant: Huawei Technologies Co., Ltd.

Inventor： Changzheng ZHANG ,  Xiaolong BAI ,  Dandan TU

IPC: G06N3/08 ,  G06N3/04

Abstract: A method for training a neural network model are disclosed. Each training period includes K iterations, and for an ith iteration of one of N worker modules within each training period, each worker module performs in parallel the following steps: calculating a model parameter of an (i+1)th iteration based on a local gradient of the ith iteration and a model parameter of the ith iteration, and if i is less than K, calculating a local gradient of the (i+1)th iteration based on the model parameter of the (i+1)th iteration and sample data of the (i+1)th iteration; and pulling, by the worker module, a global gradient of an rth iteration from a server module and/or pushing, by the worker module, a local gradient of an fth iteration to the server module. In this way, time windows of a calculation process and a communication process overlap, thereby reducing time delay.