-
1.发明公开公开(公告)号：US20240354582A1
公开(公告)日：2024-10-24
申请号：US18441076
申请日：2024-02-14
发明人： Seon-Tae KIM , Joon-Sun AUH , Chang-Sik CHO
IPC分类号： G06N3/0895
CPC分类号： G06N3/0895
摘要： Disclosed herein are a method for contrastive learning of a neural network for automatic data labeling and an apparatus for the same. The method includes generating transformed data samples for unlabeled input data samples corresponding to a batch size, detecting a positive sample (positive data) in the transformed data samples in consideration of the similarity between a single data sample for contrastive learning, among the input data samples, and each of the transformed data samples, and performing contrastive learning of a neural network based on the loss value of the positive sample.
-
2.发明公开公开(公告)号：US20240069875A1
公开(公告)日：2024-02-29
申请号：US18209617
申请日：2023-06-14
发明人： Jae-Bok PARK , Chang-Sik CHO , Kyung-Hee LEE , Ji-Young KWAK , Seon-Tae KIM , Hong-Soog KIM , Jin-Wuk SEOK , Hyun-Woo CHO
摘要： Disclosed herein are a neural network model deployment method and apparatus for providing a deep learning service. The neural network model deployment method may include providing a specification wizard to a user, searching for and training a neural network based on a user requirement specification that is input through the specification wizard, generating a neural network template code based on the user requirement specification and the trained neural network, converting the trained neural network into a deployment neural network that is usable in a target device based on the user requirement specification, and deploying the deployment neural network to the target device.
-
公开(公告)号：US20230342133A1
公开(公告)日：2023-10-26
申请号：US18098486
申请日：2023-01-18
发明人： Kyung-Hee LEE , Ji-Young KWAK , Seon-Tae KIM , Jae-Bok PARK , Ik-Soo SHIN , Chang-Sik CHO
摘要： Disclosed herein are an apparatus and method for generating a neural network executable image. The apparatus includes one or more processors and executable memory for storing at least one program executed by the one or more processors. The at least one program receives user requirements including a default neural network model and training result data for generating a neural network executable image required by a user, checks whether the default neural network model included in the user requirements is capable of being supported in a target system in which the neural network executable image is to be installed, converts the default neural network model into a neural network model executable in the target system, converts the training result data by reconfiguring the data format set of the training result data, and generates a neural network executable image by combining the converted neural network model and the converted training result data.
-
公开(公告)号：US20230316091A1
公开(公告)日：2023-10-05
申请号：US18166948
申请日：2023-02-09
发明人： Jin-Wuk SEOK , Ji-Young KWAK , Seon-Tae KIM , Hong-Soog KIM , Jae-Bok PARK , Kyung-Hee LEE , Chang-Sik CHO , Hyun-Woo CHO
IPC分类号： G06N3/098
CPC分类号： G06N3/098
摘要： Disclosed herein are a federated learning method and apparatus. The federated learning method includes receiving a feature vector extracted from a client side and label data corresponding to the feature vector, outputting a feature vector with phase information preserved therein by applying the feature vector as input of a Self-Organizing Feature Map (SOFM), and training a neural network model by applying both the feature vector with the phase information preserved therein and the label data as input of a neural network model.
-
公开(公告)号：US20170199763A1
公开(公告)日：2017-07-13
申请号：US15206767
申请日：2016-07-11
发明人： Yu Seung MA , Kyeong Tae KIM , Bum Ho KIM , Sang Cheol KIM , Seon-Tae KIM , Hae Yong KIM , Pyeong Soo MAH , Jun Keun SONG , Gyu Sang SHIN , Duk Kyun WOO , Jeong-Woo LEE
CPC分类号： G06F9/4881 , G06F11/323 , G06F11/3404 , G06F16/24578 , G06F16/26 , G06F2209/483
摘要： A method and an apparatus for visualizing a scheduling result in a multicore system. A method for visualizing a scheduling result of a plurality of tasks with respect to a plurality of cores in a multicore system includes extracting scheduling data in a time section to be visualized, determining whether the number of the extracted scheduling data exceeds a preset first threshold value, if the number of the extracted scheduling data exceeds the preset first threshold value, performing reduction of the extracted scheduling data, and visualizing and outputting the reduced scheduling data.
-
6.发明申请APPARATUS AND METHOD FOR SETTING UP ACTIVE NETWORKING OF SMART DEVICES FOR PROVIDING CONVERGED SERVICE 审中-公开用于设置用于提供转换服务的智能设备的主动联网的装置和方法公开(公告)号：US20140330905A1
公开(公告)日：2014-11-06
申请号：US14138384
申请日：2013-12-23
发明人： Ji-Young KWAK , Seon-Tae KIM
IPC分类号： H04L29/08
CPC分类号： H04W84/18 , H04L45/026 , H04L45/124 , H04L65/1006 , H04L65/1013 , H04L67/02 , H04L67/10 , H04L67/1042 , H04L67/1044 , H04L67/1093 , H04L67/12 , H04W4/02 , H04W40/10 , H04W76/14
摘要： Disclosed herein is a method for setting up active networking of smart devices for providing a converged service. In the method of setting up active networking of smart devices for providing a converged service, each of a plurality of smart objects broadcasts location information thereof. A smart object neighboring a largest number of smart objects is selected as a smart zone manager from among the plurality of smart objects. The selected smart zone manager broadcasts smart zone information to neighbor smart objects, and then forms a smart zone.
摘要翻译： 这里公开了一种用于设置用于提供融合服务的智能设备的主动联网的方法。 在设置用于提供融合服务的智能设备的主动组网的方法中，多个智能对象中的每一个广播其位置信息。 从多个智能对象中选择与最大数量的智能对象相邻的智能对象作为智能区管理器。 所选择的智能区管理器将智能区信息广播到邻居智能对象，然后形成智能区。
-
7.发明申请公开(公告)号：US20140313967A1
公开(公告)日：2014-10-23
申请号：US14252204
申请日：2014-04-14
发明人： Woo-Suk CHA , Seon-Tae KIM , Jun-Keun SONG , Mi-Sun YU , Ji-Young KWAK
IPC分类号： H04W4/06
CPC分类号： H04W4/06
摘要： A multicast routing apparatus in a wireless mesh network and a method thereof are disclosed. The multicast routing apparatus in the wireless mesh network includes: an information obtaining unit configured to obtain group identification (ID) information of neighboring mesh stations corresponding to the respective mesh stations; a selecting unit configured to select a multicast forwarding candidate corresponding to a candidate of the mesh stations capable of participating in forwarding multicast data packet using the group ID information; a tree generating unit configured to generate a tree corresponding to a transmission path of the multicast data packet; and a packet transmitting unit configured to transmit data packet from a transmitting mesh station to a receiving mesh station of the mesh stations using the tree.
摘要翻译： 公开了无线网状网络中的组播路由装置及其方法。 无线网状网络中的组播路由装置包括：信息获取单元，被配置为获取对应于各个网格站的相邻网格站的组标识（ID）信息; 选择单元，被配置为使用所述组ID信息来选择与能够参与转发多播数据分组的网格站的候选对应的组播转发候选; 树生成单元，被配置为生成与所述多播数据分组的传输路径对应的树; 以及分组发送单元，被配置为使用该树将来自发送网状站的数据分组发送到网状站的接收网状站。
-
公开(公告)号：US20230315402A1
公开(公告)日：2023-10-05
申请号：US18098402
申请日：2023-01-18
发明人： Chang-Sik CHO , Jae-Bok PARK , Kyung-Hee LEE , Ji-Young KWAK , Seon-Tae KIM , Ik-Soo SHIN
摘要： Disclosed herein are an apparatus and method for developing a neural network application. The apparatus includes one or more processors and executable memory for storing at least one program executed by the one or more processors. The at least one program receives a target specification and an application specification including user requirements, searches for a neural network model corresponding to the target specification and the application specification in a database, builds an inference engine for performing a neural network operation used by the neural network model, and generates a target image for executing the neural network model to be suitable for a target device using the inference engine.
-
公开(公告)号：US20190037578A1
公开(公告)日：2019-01-31
申请号：US16044330
申请日：2018-07-24
发明人： Kyeong-Tae KIM , Bum-Ho KIM , Sang-Cheol KIM , Seon-Tae KIM , Hae-Yong KIM , Yu-Seung MA , Pyeong-Soo MAH , Hui-Ung PARK , Duk-Kyun WOO , Jeong-Woo LEE
摘要： Disclosed herein are a dynamic scheduling method for guaranteeing Quality of Service QoS depending on network transmission traffic and a system using the same. The dynamic scheduling method includes assigning communication channels to respective nodes based on Identifications (IDs) of parent nodes corresponding to the respective nodes, setting priorities for assignment of time slots to the respective nodes in each quarter based on data traffic volumes corresponding to the respective nodes, and assigning time slots to the respective nodes in each quarter depending on the set priorities for assignment of the time slots.
-
公开(公告)号：US20170257814A1
公开(公告)日：2017-09-07
申请号：US15207417
申请日：2016-07-11
发明人： Kyeong Tae KIM , Bum Ho KIM , Sang Cheol KIM , Seon-Tae KIM , Hae Yong KIM , Yu Seung MA , Pyeong Soo MAH , Gyu Sang SHIN , Duk Kyun WOO , Jeong-Woo LEE
CPC分类号： H04W40/244 , H04W8/005 , H04W24/02 , H04W40/12 , H04W48/12 , H04W72/0446 , H04W76/10 , H04W84/20
摘要： There is provided a network join method including transmitting a long beacon message including transmission timing information of a first short beacon message to a child node network device, transmitting a first short beacon message including information regarding an interval in which it is possible to transmit a message, to the child node network device according to a transmission timing of the first short beacon message, receiving a slot allocation request message from the child node network device according to the interval in which it is possible to transmit a message, and checking whether the child node network device joins a network, and transmitting a slot allocation confirmation message to the child node network device.
-
-
-
-
-
-
-
-
-
搜索结果
12 条记录 (耗时 0.017 秒)
