公开(公告)号：US20240311651A1

公开(公告)日：2024-09-19

申请号：US18668637

申请日：2024-05-20

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Binxin RU ,  Xingchen WAN ,  Pedro ESPERANCA ,  Fabio Maria CARLUCCI ,  Zhenguo LI

IPC: G06N3/0985 ,  G06N3/04

CPC classification number: G06N3/0985 ,  G06N3/04

Abstract: Disclosed is a method for searching for a neural network architecture ensemble model. The method includes: obtaining a dataset, where the dataset includes a sample and an annotation in a classification task; performing search by using a distributional neural network architecture search algorithm, including: determining a hyperparameter of a neural network architecture distribution; sampling a valid neural network architecture from the architecture distribution defined by the hyperparameter; training and evaluating the neural network architecture on the dataset, to obtain a performance indicator; determining, based on the performance indicator, neural network architecture distributions that share the hyperparameter, to obtain a candidate pool of base learners; and determining a surrogate model; and predicting test performance of the base learner in the candidate pool by using the surrogate model, and determining that k diverse base learners that meet a task scenario requirement form an ensemble model.

公开(公告)号：US20220108546A1

公开(公告)日：2022-04-07

申请号：US17553236

申请日：2021-12-16

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Hang XU ,  Zhenguo LI

IPC: G06V10/82 ,  G06V10/764 ,  G06V10/77 ,  G06N3/08 ,  G06N5/02

Abstract: This application provides an object detection method and apparatus. This application relates to the field of artificial intelligence, and specifically, to the field of computer vision. The method includes: obtaining a to-be-detected image; performing convolution processing on the to-be-detected image to obtain an initial image feature of a to-be-detected object in the to-be-detected image; determining an enhanced image feature of the to-be-detected object based on knowledge graph information; and determining a candidate frame and a classification of the to-be-detected object based on the initial image feature and the enhanced image feature of the to-be-detected object. The enhanced image feature indicates semantic information of a different object category corresponding to another object associated with the to-be-detected object. Therefore, in this application, an effect of the object detection method can be improved.

公开(公告)号：US20230048405A1

公开(公告)日：2023-02-16

申请号：US17975436

申请日：2022-10-27

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Yimin HUANG ,  Yujun LI ,  Zhenguo LI

IPC: G06V10/776 ,  G06V10/82

Abstract: The present disclosure relates to neural network optimization methods and apparatuses in the field of artificial intelligence. One example method includes sampling preset hyperparameter search space to obtain multiple hyperparameter combinations. Multiple iterative evaluations are performed on the multiple hyperparameter combinations to obtain multiple performance results of each hyperparameter combination. Any iterative evaluation comprises obtaining at least one performance result of each hyperparameter combination, and if a hyperparameter combination meets a first preset condition, re-evaluating the hyperparameter combination to obtain a re-evaluated performance result of the hyperparameter combination. An optimal hyperparameter combination is determined. If the optimal hyperparameter combination does not meet a second preset condition, a preset model is updated, based on the multiple performance results of each hyperparameter combination, for next sampling. Or if the optimal hyperparameter combination meets a second preset condition, the optimal hyperparameter combination is used as a hyperparameter combination of a neural network.

公开(公告)号：US20220092351A1

公开(公告)日：2022-03-24

申请号：US17538640

申请日：2021-11-30

Applicant: Huawei Technologies Co., Ltd. ,  PEKING UNIVERSITY

Inventor： Weiran HUANG ,  Aoxue LI ,  Zhenguo LI ,  Tiange LUO ,  Li QIAN ,  Liwei WANG

IPC: G06K9/62 ,  G06K9/00 ,  G06K9/46 ,  G06N3/08

Abstract: An image classification method, a neural network training method, and an apparatus are provided, and relate to the field of artificial intelligence, and specifically, to the field of computer vision. The image classification method includes: obtaining a to-be-processed image; and obtaining a classification result of the to-be-processed image based on a pre-trained neural network model, where the classification result includes a class or a superclass to which the to-be-processed image belongs. When the neural network model is trained, not only labels of a plurality of training images but also class hierarchy information of the plurality of training images is used. That is, more abundant information of the training images is used. Therefore, images can be better classified.

公开(公告)号：US20200258006A1

公开(公告)日：2020-08-13

申请号：US16863110

申请日：2020-04-30

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Fei CHEN ,  Zhenhua DONG ,  Zhenguo LI ,  Xiuqiang HE ,  Li QIAN ,  Shuaihua PENG

IPC: G06N20/00 ,  G06N5/02

Abstract: Example prediction methods and apparatus are described. One example includes sending a first model parameter and a second model parameter by a server to a plurality of terminals. The first model parameter and the second model parameter are adapted to a prediction model of the terminal. The server receives a first prediction loss sent by at least one of the plurality of terminals. A first prediction loss sent by each of the at least one terminal is calculated by the terminal based on the prediction model that uses the first model parameter and the second model parameter. The server updates the first model parameter based on the first prediction loss sent by the at least one terminal to obtain an updated first model parameter. The server updates the second model parameter based on the first prediction loss sent by the at least one terminal to obtain an updated second model parameter.

公开(公告)号：US20240135191A1

公开(公告)日：2024-04-25

申请号：US18392502

申请日：2023-12-21

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Mi LUO ,  Fei CHEN ,  Zhenguo LI ,  Jiashi FENG

IPC: G06N3/098 ,  G06N3/082

CPC classification number: G06N3/098 ,  G06N3/082

Abstract: A method, an apparatus, and a system for generating a neural network model, a device, a medium, and a program product are provided. In an embodiment, a first device sends an indication about a structure of a subnetwork model to a second device, where the subnetwork model is determined by adjusting a structure of a hypernetwork model. The first device receives a parameter of the subnetwork model from the second device, where the parameter of the subnetwork model is determined by the second device based on the indication and the hypernetwork model. The first device trains the subnetwork model based on the received parameter of the subnetwork model. The first device sends a parameter of the trained subnetwork model to the second device for the second device to update the hypernetwork model. In the foregoing manner, an efficient federated learning scheme between a plurality of devices is provided.

公开(公告)号：US20230141145A1

公开(公告)日：2023-05-11

申请号：US18150748

申请日：2023-01-05

Applicant: Huawei Technologies Co., Ltd.

Inventor： Weijun HONG ,  Guilin LI ,  Weinan ZHANG ,  Yong YU ,  Xing ZHANG ,  Zhenguo LI

IPC: G06N3/04 ,  G06N3/082

CPC classification number: G06N3/04 ,  G06N3/082

Abstract: A neural network building method and apparatus are disclosed, and relate to the field of artificial intelligence. The method includes: initializing a search space and a plurality of building blocks, where the search space includes a plurality of operators, and the building block is a network structure obtained by connecting a plurality of nodes by using the operator; during training, in at least one training round, randomly discarding some operators, and updating the plurality of building blocks by using operators that are not discarded; and building a target neural network based on the plurality of updated building blocks. In the method, some operators are randomly discarded. This breaks association between operators, and overcomes a co-adaptation problem during training, to obtain a target neural network with better performance.

公开(公告)号：US20190286986A1

公开(公告)日：2019-09-19

申请号：US16431393

申请日：2019-06-04

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Wu BIN ,  Fengwei ZHOU ,  Zhenguo LI

IPC: G06N3/08

Abstract: Embodiments of the present invention provide a machine learning model training method, including: obtaining target task training data and N categories of support task training data; inputting the target task training data and the N categories of support task training data into a memory model to obtain target task training feature data and N categories of support task training feature data; training the target task model based on the target task training feature data and obtaining a first loss of the target task model, and separately training respectively corresponding support task models based on the N categories of support task training feature data and obtaining respective second losses of the N support task models; and updating the memory model, the target task model, and the N support task models based on the first loss and the respective second losses of the N support task models.

公开(公告)号：US20240232575A1

公开(公告)日：2024-07-11

申请号：US18618100

申请日：2024-03-27

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Xingchen WAN ,  Binxin RU ,  Pedro ESPERANCA ,  Fabio Maria CARLUCCI ,  Zhenguo LI

IPC: G06N3/04

CPC classification number: G06N3/04

Abstract: A neural network obtaining method, a data processing method, and a related device are disclosed. The disclosed methods may be used in the field of automatic neural architecture search technologies in the field of artificial intelligence. An example method includes: obtaining first indication information, where the first indication information indicates a probability and/or a quantity of times that k neural network modules appear in a first neural architecture cell; generating the first neural architecture cell based on the first indication information, and generating a first neural network; obtaining a target score corresponding to the first indication information, where the target score indicates performance of the first neural network; and obtaining second indication information from a plurality of pieces of first indication information based on a plurality of target scores, and obtaining a target neural network corresponding to the second indication information.

公开(公告)号：US20230306077A1

公开(公告)日：2023-09-28

申请号：US18327584

申请日：2023-06-01

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Huifeng GUO ,  Bo CHEN ,  Ruiming TANG ,  Zhenguo LI ,  Xiuqiang HE

IPC: G06F17/18

CPC classification number: G06F17/18

Abstract: Embodiments of this application provide a data processing method and apparatus to better learn a vector representation value of each feature value in a continuous feature. The method specifically includes: The data processing apparatus obtains the continuous feature from sample data, and then performs discretization processing on the continuous feature by using a discretization model, to obtain N discretization probabilities corresponding to the continuous feature. The N discretization probabilities correspond to N preset meta-embeddings, and N is an integer greater than 1. Finally, the data processing apparatus determines a vector representation value of the continuous feature based on the N discretization probabilities and the N meta-embeddings.