-
公开(公告)号：US20210406767A1
公开(公告)日：2021-12-30
申请号：US17142822
申请日：2021-01-06
Inventor： Daxiang DONG , Weibao GONG , Yi LIU , Dianhai YU , Yanjun MA , Haifeng WANG
IPC: G06N20/00 , G06F16/182 , G06N5/04
Abstract: The present application discloses a distributed training method and system, a device and a storage medium, and relates to technical fields of deep learning and cloud computing. The method includes: sending, by a task information server, a first training request and information of an available first computing server to at least a first data server; sending, by the first data server, a first batch of training data to the first computing server, according to the first training request; performing, by the first computing server, model training according to the first batch of training data, sending model parameters to the first data server so as to be stored after the training is completed, and sending identification information of the first batch of training data to the task information server so as to be recorded; wherein the model parameters are not stored at any one of the computing servers.
-
12.发明申请公开(公告)号：US20210294975A1
公开(公告)日：2021-09-23
申请号：US17015411
申请日：2020-09-09
Inventor： Xinchao XU , Haifeng WANG , Hua WU , Zhanyi LIU
IPC: G06F40/284 , G10L15/06 , G06N3/08 , G10L15/04 , G06F40/274
Abstract: A method, an electronic device and a readable storage medium for creating a label marking model are disclosed. According to an embodiment, the method for creating the label marking model includes: obtaining text data and determining a word or phrase to be marked in the text data; according to the word or phrase to be marked, constructing a first training sample of the text data corresponding to a word or phrase replacing task and a second training sample corresponding to a label marking task; training a neural network model with a plurality of the first training samples and a plurality of the second training samples, respectively, until a loss function of the word or phrase replacing task and a loss function of the label marking task satisfy a preset condition, to obtain the label marking model. The technical solution may improve the accuracy of the label marking model.
-
公开(公告)号：US20210209417A1
公开(公告)日：2021-07-08
申请号：US17209576
申请日：2021-03-23
Inventor： Daxiang DONG , Wenhui ZHANG , Zhihua WU , Dianhai YU , Yanjun MA , Haifeng WANG
Abstract: A method and an apparatus for generating a shared encoder are provided, which belongs to a field of computer technology and deep learning. The method includes: sending by a master node a shared encoder training instruction to child nodes, so that each child node obtains training samples based on a type of a target shared encoder included in the training instruction; sending an initial parameter set of the target shared encoder to be trained to each child node after obtaining a confirmation message returned by each child node; obtaining an updated parameter set of the target shared encoder returned by each child node; determining a target parameter set corresponding to the target shared encoder based on a first preset rule and the updated parameter set of the target shared encoder returned by each child node.
-
公开(公告)号：US20210034993A1
公开(公告)日：2021-02-04
申请号：US16936190
申请日：2020-07-22
Inventor： Miao FAN , Jizhou HUANG , An ZHUO , Ying LI , Ping LI , Haifeng WANG
Abstract: A POI valuation method, apparatus, device and computer storage medium are disclosed. The method comprises: obtaining information of first POIs with known values and information of second POIs with unknown values within a regional range; creating a valuation model which is configured to revaluate a first POI using values of surrounding POIs of the first POI, the surrounding POIs including other first POIs and second POIs within a predetermined range of distance from the first POI, and adjusting values of second POIs in the surrounding POIs using an error between a revaluated value of first POI and the known value of the first POI; training the valuation model until the error is minimized; obtaining the values of the second POIs from the valuation model. The solutions may reduce the requirement for manpower and improve the valuation efficiency as compared with manually valuation of POIs one by one.
-
公开(公告)号：US20180357570A1
公开(公告)日：2018-12-13
申请号：US16006208
申请日：2018-06-12
Inventor： Ke SUN , Shiqi ZHAO , Dianhai YU , Haifeng WANG
CPC classification number: G06N99/005 , G06F7/14 , G06F17/2785
Abstract: A method and apparatus for building a conversation understanding system based on artificial intelligence, a device and a computer-readable storage medium. In embodiments of the present disclosure, it is feasible to obtain the training feedback information provided by conversation service conducted by the user and the basic conversation understanding system, then according to the training feedback information, perform adjustment processing for a service state of the basic conversation understanding system, to obtain an adjustment state of the basic conversation understanding system. It is possible to perform data merging processing according to the training feedback information and the adjustment state of the basic conversation understanding system, to obtain model training data for building the model conversation understanding system. This method does not require persons to participate in annotation operations of the training data, exhibits simple operations and a high correctness rate, improving the efficiency and reliability of the conversation understanding system.
-
Patent Agency Ranking
公开(公告)号：US20170228459A1
公开(公告)日：2017-08-10
申请号：US15384141
申请日：2016-12-19
Inventor： Haifeng WANG , Shiqi ZHAO , Haifeng WU , Tian WU , Daisong GUAN
CPC classification number: G06F16/951 , G06F16/9535 , G06N20/00
Abstract: A method and a device for mobile searching based on artificial intelligence are provided in the present disclosure. The method includes: displaying a search box, and receiving a query inputted by a user via the search box; obtaining a search result according to the query, and displaying the search result on a search result page; after receiving a click instruction on the search result, displaying a context page corresponding to the search result; and after receiving a click instruction on a result in the search result or in the context page, displaying a content page corresponding to the result clicked. The method can break through the concept of PC search and provide a search method which is more suitable for a mobile search scene.
-
公开(公告)号：US20220286521A1
公开(公告)日：2022-09-08
申请号：US17824318
申请日：2022-05-25
Inventor： Yunpeng PENG , Haifeng WANG
IPC: H04L67/306 , H04L67/50 , G06K9/62
Abstract: An object recommendation method, computer device, and medium are provided, relating to the field of artificial intelligence and, particularly, content recommendation. A method includes: obtaining a first user profile of a user, the first user profile being determined based on behavior data of the user over a first historical period of time; using a matching model to determine a recommended object based on the first user profile; recommending the recommended object to the user; obtaining a second user profile of the user, the second user profile being determined based on behavior data of the user over a second historical period of time, and the behavior data of the user over the second historical period of time includes behavior data of the user after the recommended object is recommended to the user; and updating the matching model based on the first user profile, the second user profile, and the recommended object.
-
公开(公告)号：US20210383797A1
公开(公告)日：2021-12-09
申请号：US17411917
申请日：2021-08-25
Inventor： Fan WANG , Siqi BAO , Huang HE , Hua WU , Jingzhou HE , Haifeng WANG
Abstract: A method for dialogue processing, an electronic device and a storage medium are provided. The specific technical solution includes: obtaining a dialogue history; selecting a target machine from a plurality of machines; inputting the dialogue history into a trained dialogue model in the target machine to generate a response to the dialogue history, in which the dialogue model comprises a common parameter and a specific parameter, and different machines correspond to the same common parameter.
-
公开(公告)号：US20210383064A1
公开(公告)日：2021-12-09
申请号：US17101789
申请日：2020-11-23
Inventor： Shuohuan WANG , Siyu DING , Yu SUN , Hua WU , Haifeng WANG
IPC: G06F40/279 , G06F40/166 , G06F40/30 , G06N20/00
Abstract: The disclosure provides a text recognition method, an electronic device, and a storage medium. The method includes: obtaining N segments of a sample text; inputting each of the N segments into a preset initial language model in sequence, to obtain first text vector information corresponding to the N segments; inputting each of the N segments into the initial language model in sequence again, to obtain second text vector information corresponding to a currently input segment; in response to determining that the currently input segment has the mask, predicting the mask according to the second text vector information and the first text vector information to obtain a predicted word at a target position corresponding to the mask; training the initial language model according to an original word and the predicted word to generate a long text language model; and recognizing an input text through the long text language model.
-
20.发明申请公开(公告)号：US20210357814A1
公开(公告)日：2021-11-18
申请号：US17362674
申请日：2021-06-29
Inventor： Xinxuan WU , Xuefeng YAO , Dianhai YU , Zhihua WU , Yanjun MA , Tian WU , Haifeng WANG
IPC: G06N20/00
Abstract: The present disclosure provides a method and apparatus for distributed training a model, an electronic device, and a computer readable storage medium. The method may include: performing, for each batch of training samples acquired by a distributed first trainer, model training through a distributed second trainer to obtain gradient information; updating a target parameter in a distributed built-in parameter server according to the gradient information; and performing, in response to determining that training for a preset number of training samples is completed, a parameter exchange between the distributed built-in parameter server and a distributed parameter server through the distributed first trainer to perform a parameter update on the initial model until training for the initial model is completed.
-
-
-
-
-
-
-
-
-
Search Results
30
records
(took 0.032 seconds)
