公开(公告)号：US20230229740A1

公开(公告)日：2023-07-20

申请号：US18070979

申请日：2022-11-29

Applicant: Electronics and Telecommunications Research Institute

Inventor： MINHO PARK ,  Dong-oh KANG ,  Hwajeon SONG ,  Jeun Woo LEE

IPC: G06F18/2431 ,  G06F18/28

CPC classification number: G06F18/2431 ,  G06F18/28

Abstract: The present invention provides a multiclass classification apparatus and method robust to imbalanced data, which generate artificial data of a minority class on the basis of an over-sampling technique based on adversarial learning to balance imbalanced data and performs multiclass classification robust to imbalanced data by using corresponding data in class classification learning without additionally collecting data.

公开(公告)号：US20140192653A1

公开(公告)日：2014-07-10

申请号：US13769775

申请日：2013-02-18

Applicant: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE

Inventor： Joon Young JUNG ,  Dong-oh KANG ,  Chang Seok BAE

IPC: H04L29/08

CPC classification number: H04L67/1002

Abstract: An apparatus for controlling load allocation in a cluster system includes a cluster module having a plurality of target nodes and a cluster power management module. The cluster power management module analyzes resource usage of the target nodes by monitoring the load states of the target nodes. The cluster power management module adaptively allocates loads to the target nodes based on the analyzed resource usage and N allocation thresholds settled in response to an increase of the analyzed resource usage. The cluster power management module controls the target nodes so that the power state can be changed with the adaptively allocated loads.

Abstract translation: 用于控制集群系统中的负载分配的装置包括具有多个目标节点的集群模块和集群电源管理模块。 集群电源管理模块通过监视目标节点的负载状态来分析目标节点的资源使用情况。 集群电源管理模块基于所分析的资源使用和响应于分析的资源使用的增加而确定的N个分配阈值自适应地将负载分配给目标节点。 集群电源管理模块控制目标节点，以便通过自适应分配的负载来改变电源状态。

公开(公告)号：US20230281477A1

公开(公告)日：2023-09-07

申请号：US18103962

申请日：2023-01-31

Applicant: Electronics and Telecommunications Research Institute

Inventor： Chun-Hee LEE ,  Dong-oh KANG ,  Hwa Jeon SONG

IPC: G06N5/022 ,  G06N5/02

CPC classification number: G06N5/022 ,  G06N5/027

Abstract: A learning method for improving performance of a knowledge graph embedding model is provided. The method includes: performing learning of a first knowledge graph embedding model based on input knowledge data; extracting all embedding vectors from the learned first knowledge graph embedding model, and extracting prior knowledge based on the extracted embedding vectors; and performing learning of a second knowledge graph embedding model through at least one of initialization of the embedding vectors and transform of the input knowledge data based on the extracted prior knowledge.