公开(公告)号：US20240152805A1

公开(公告)日：2024-05-09

申请号：US18384634

申请日：2023-10-27

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Hao LI ,  Gopi Krishnan Rajbahadur ,  Dayi Lin ,  Zhenming Jiang

IPC: G06N20/00

CPC classification number: G06N20/00

Abstract: A method for detecting and/or preventing overfitting in training of deep learning and neural network models. The method has a classifier-training method, an overfitting-detection method, and an overfitting-prevention method. The classifier-training method trains one or more classifiers using training histories and labels of one or more trained machine-learning (ML) models. The overfitting-detection method uses the trained classifiers based on the training history such as validation losses of a trained target ML model to identify an overfitting status of the trained target ML model. The overfitting-prevention method is performed during the training of a target ML model and uses the trained classifiers based on the training history of the target ML model to identify and preventing overfitting of the target ML model.

公开(公告)号：US20240152578A1

公开(公告)日：2024-05-09

申请号：US18386023

申请日：2023-11-01

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Xu Yang ,  Gopi Krishnan Rajbahadur ,  Dayi Lin

IPC: G06F18/22 ,  G06T11/20

CPC classification number: G06F18/22 ,  G06T11/206

Abstract: A computerized method for detecting and analyzing data clones in one or more dataset pairs has the steps of: obtaining one or more similarity matrices and one or more sets of readout values of the one or more similarity matrices from the dataset pairs using a data-clone detection method, each set of readout values corresponding to a similarity matrix; obtaining one or more importance values for the one or more similarity matrices by processing the one or more sets of readout values using an interpretation method, each importance value corresponding to a similarity matrix; obtaining one or more weighted similarity matrices by weighting each similarity matrix using the corresponding importance value; and obtaining one or more summed similarity matrices by grouping and summing the weighted similarity matrices according to one or more categories for providing a result with indications of locations of the data clones in the dataset pairs.