公开(公告)号：US20250007792A1

公开(公告)日：2025-01-02

申请号：US18828823

申请日：2024-09-09

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Chujie Wang ,  Xianming Li

IPC: H04L41/16 ,  G06N5/022 ,  H04L41/082 ,  H04L41/0894

Abstract: A knowledge update method performed by a knowledge provision network element includes sending knowledge information indicating quality of intent knowledge of a first intent operation to a network management network element. Policy information indicating a manner of updating the intent knowledge is received from the network management network element, and the intent knowledge is updated based on the policy information. In this way, the knowledge provision network element can feed back the quality of the intent knowledge to the network management network element, the network management network element can deliver the manner of updating the intent knowledge to the knowledge provision network element based on the quality, and the knowledge provision network element can perform updating based on the policy information in a subsequent intent knowledge update procedure, to improve quality of updated knowledge.

公开(公告)号：US20230274192A1

公开(公告)日：2023-08-31

申请号：US18314802

申请日：2023-05-09

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Chujie Wang ,  Dongrun Qin ,  Yijun Yu

IPC: G06N20/00

CPC classification number: G06N20/00

Abstract: A federated learning method, apparatus, and system are disclosed. A first node obtains data distribution information of a plurality of second nodes based on a target data feature required by a training task; the first node selects at least two target second nodes from the plurality of second nodes based on a target data class required by the training task and the data distribution information of the plurality of second nodes; and the first node indicates the at least two target second nodes to perform federated learning, to obtain a federated learning model that is in the training task and that corresponds to the target data class. In this way, when participants have a plurality of data distributions, a trained model is prevented, as much as possible, from being affected by data poisoning.

公开(公告)号：US20240356798A1

公开(公告)日：2024-10-24

申请号：US18758912

申请日：2024-06-28

Applicant: Huawei Technologies Co., Ltd.

Inventor： Chujie Wang ,  Xianming Li

IPC: H04L41/0677 ,  H04L41/0631

CPC classification number: H04L41/0677 ,  H04L41/0631

Abstract: Embodiments of this application provide a communication method and a communication apparatus. The communication method includes: A first network element sends at least one piece of check content to a second network element, where the check content indicates an abnormal event that causes a fault of a second network; and the first network element receives at least one check result from the second network element, where the check result indicates whether the abnormal event occurs in the second network. In this way, the first network element can obtain the check result detected by the second network element based on the check content, to determine a fault root cause. This implements cross-operator fault diagnosis, and improves a fault diagnosis capability and an intent achievement rate.