-
公开(公告)号：US20230237309A1
公开(公告)日：2023-07-27
申请号：US18180841
申请日：2023-03-08
发明人： Xiaoyun Zhou , Jiacheng Sun , Nanyang Ye , Xu Lan , Qijun Luo , Pedro Esperanca , Fabio Maria Carlucci , Zewei Chen , Zhenguo Li
摘要： A device for machine learning is provided, including a first neural network layer, a second neural network layer with a normalization layer arranged in between. The normalization layer is configured to, when the device is undergoing training on a batch of training samples, receive multiple outputs of the first neural network layer for a plurality of training samples of the batch, each output comprising multiple data values for different indices on a first dimension and a second dimension; group the outputs into multiple groups based on the indices on the first and second dimensions; form a normalization output for each group which are provided as input to the second neural network layer. According to the application, the training of a deep convolutional neural network with good performance that performs stably at different batch sizes and is generalizable to multiple vision tasks is achieved, thereby improving the performance of the training.
-
2.发明授权公开(公告)号：US11334758B2
公开(公告)日：2022-05-17
申请号：US16729043
申请日：2019-12-27
发明人： Ruiming Tang , Huifeng Guo , Zhenguo Li , Xiuqiang He
IPC分类号： G06K9/62
摘要： The method includes: obtaining a plurality of pieces of feature data; automatically performing two different types of nonlinear combination processing operations on the plurality of pieces of feature data to obtain two groups of processed data, where the two groups of processed data include a group of higher-order data and a group of lower-order data, the higher-order data is related to a nonlinear combination of m pieces of feature data in the plurality of pieces of feature data, and the lower-order data is related to a nonlinear combination of n pieces of feature data in the plurality of pieces of feature data, where m≥3, and m>n≥2; and determining prediction data based on a plurality of pieces of target data, where the plurality of pieces of target data include the two groups of processed data.
-
公开(公告)号：US20170366197A1
公开(公告)日：2017-12-21
申请号：US15697066
申请日：2017-09-06
发明人： Zhenguo Li , Ge Luo , Ke Yi , Wei Fan , Cheng He
IPC分类号： H03M7/30
CPC分类号： H03M7/30 , H03M7/3064
摘要： A method for compressing flow data, including: generating multiple line segments according to flow data and a predefined maximum error that are acquired; obtaining a target piecewise linear function according to the multiple line segments, where the target piecewise linear function includes multiple linear functions, and an intersection set of value ranges of independent variables of every two linear functions among the multiple linear functions includes a maximum of one value; and outputting a reference data point according to the target piecewise linear function, where the reference data point includes a point of continuity and a point of discontinuity of the target piecewise linear function. In this way, a maximum error, a target piecewise linear function is further determined according to the multiple line segments, and a point of continuity and a point of discontinuity of the target piecewise linear function are used to represent compressed flow data.
-
公开(公告)号：US20230206069A1
公开(公告)日：2023-06-29
申请号：US18175936
申请日：2023-02-28
发明人： Junlei Zhang , Chuanjian Liu , Guilin Li , Xing Zhang , Wei Zhang , Zhenguo Li
摘要： A deep learning training method includes obtaining a training set, a first neural network, and a second neural network, where shortcut connections included in the first neural network are less than shortcut connections included in the second neural network; performing at least one time of iterative training on the first neural network based on the training set, to obtain a trained first neural network, where any iterative training includes: using a first output of at least one first intermediate layer in the first neural network as an input of at least one network layer in the second neural network, to obtain an output result of the at least one network layer; and updating the first neural network according to a first loss function.
-
公开(公告)号：US10579703B2
公开(公告)日：2020-03-03
申请号：US15694559
申请日：2017-09-01
发明人： Zhenguo Li , Jiefeng Cheng , Wei Fan
IPC分类号： G06F17/12 , G06F16/951 , G06F7/00 , G06F17/16 , G06F17/10 , G06F16/903 , G06F7/02 , G06Q30/02
摘要： A similarity measurement method includes: obtaining a directional relationship between nodes in a network, and determining a transition matrix; calculating a constraint matrix according to the transition matrix and an obtained attenuation factor; constructing a system of linear equations, where a coefficient matrix of the system of linear equations is the constraint matrix, and a variable of the system of linear equations is a correction vector; solving the system of linear equations by means of iteration by using a Jacobi method, and determining the correction vector; and calculating similarities between the nodes according to the transition matrix, the attenuation factor, and a diagonal correction matrix that is generated according to the correction vector. In the method, the correction vector is determined by using the Jacobi method, and further the similarities between the nodes may be calculated.
-
公开(公告)号：US10218381B2
公开(公告)日：2019-02-26
申请号：US15697066
申请日：2017-09-06
发明人： Zhenguo Li , Ge Luo , Ke Yi , Wei Fan , Cheng He
IPC分类号： H03M7/30
摘要： A method for compressing flow data, including: generating multiple line segments according to flow data and a predefined maximum error that are acquired; obtaining a target piecewise linear function according to the multiple line segments, where the target piecewise linear function includes multiple linear functions, and an intersection set of value ranges of independent variables of every two linear functions among the multiple linear functions includes a maximum of one value; and outputting a reference data point according to the target piecewise linear function, where the reference data point includes a point of continuity and a point of discontinuity of the target piecewise linear function. In this way, a maximum error, a target piecewise linear function is further determined according to the multiple line segments, and a point of continuity and a point of discontinuity of the target piecewise linear function are used to represent compressed flow data.
-
公开(公告)号：US20180276542A1
公开(公告)日：2018-09-27
申请号：US15993288
申请日：2018-05-30
发明人： Jiefeng Cheng , Zhenguo Li , Xiuqiang He , Dahua Lin
CPC分类号： G06N3/08 , G06F16/9535 , G06N3/04 , G06N3/0454 , G06N3/0472 , G06Q30/02 , G06Q30/0241 , G06Q30/0271
摘要： A recommendation result generation method, where the method includes obtaining article content information of at least one article and user score information of at least one user, where user score information of a first user of the at least one user includes a historical score of the first user for the at least one article, encoding the article content information and the user score information using an article neural network and a user neural network respectively to obtain a target article latent vector of each of the at least one article and a target user latent vector of each of the at least one user, and calculating a recommendation result for each user according to the article latent vector and the user latent vector.
-
公开(公告)号：US20170364478A1
公开(公告)日：2017-12-21
申请号：US15694559
申请日：2017-09-01
发明人： Zhenguo Li , Jiefeng Cheng , Wei Fan
CPC分类号： G06F17/12 , G06F7/00 , G06F7/02 , G06F16/903 , G06F16/951 , G06F17/10 , G06F17/16 , G06Q30/02
摘要： A similarity measurement method includes: obtaining a directional relationship between nodes in a network, and determining a transition matrix; calculating a constraint matrix according to the transition matrix and an obtained attenuation factor; constructing a system of linear equations, where a coefficient matrix of the system of linear equations is the constraint matrix, and a variable of the system of linear equations is a correction vector; solving the system of linear equations by means of iteration by using a Jacobi method, and determining the correction vector; and calculating similarities between the nodes according to the transition matrix, the attenuation factor, and a diagonal correction matrix that is generated according to the correction vector. In the method, the correction vector is determined by using the Jacobi method, and further the similarities between the nodes may be calculated.
-
公开(公告)号：US12067483B2
公开(公告)日：2024-08-20
申请号：US16431393
申请日：2019-06-04
发明人： Bin Wu , Fengwei Zhou , Zhenguo Li
摘要： 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.
-
10.发明授权公开(公告)号：US12039440B2
公开(公告)日：2024-07-16
申请号：US17582880
申请日：2022-01-24
发明人： Weiran Huang , Aoxue Li , Zhenguo Li , Tiange Luo , Liwei Wang
IPC分类号： G06V10/774 , G06N3/045 , G06N3/08 , G06V10/82
CPC分类号： G06N3/08 , G06N3/045 , G06V10/774 , G06V10/82
摘要： An image classification method and apparatus, and an image classification model training method and apparatus are provided, which are related to an image recognition technology in the artificial intelligence field and more specifically, to the computer vision field. The method includes: obtaining a to-be-processed image; and classifying the to-be-processed image based on a preset global class feature, to obtain a classification result of the to-be-processed image. The preset global class feature includes a plurality of class features obtained through training based on a plurality of training images in a training set. The plurality of class features in the preset global class feature are used to indicate visual features of all classes in the training set.
-
-
-
-
-
-
-
-
-
搜索结果
24 条记录 (耗时 0.024 秒)
