首页> 外国专利> MACHINE-LEARNED MODEL SWITCHING SYSTEM, EDGE DEVICE, MACHINE-LEARNED MODEL SWITCHING METHOD, AND PROGRAM

MACHINE-LEARNED MODEL SWITCHING SYSTEM, EDGE DEVICE, MACHINE-LEARNED MODEL SWITCHING METHOD, AND PROGRAM

机译:机器学习的模型切换系统,边缘设备,机器学习的模型切换方法和程序

摘要

In the field of the Internet of Things (IoT), the present invention switches, according to location, the machine-learned model that an edge device uses. A transmission means 211 of an edge device 2 transmits, to a provision device 1, location information that was generated by a positioning unit 26. When the location information is transmitted from the edge device 2 to the provision device 1, an acquisition means 111 acquires the location information and delivers the same to a specification means 112. The specification means 112 specifies a machine-learned model that corresponds to the location indicated by the location information. A provision means 113 reads, from a model DB 121, the machine-learned model specified by the specification means 112, and provides the same to the edge device 2 that transmitted the location information to the provision device 1. When the machine-learned model is provided to the edge device 2, in accordance with this provided new machine-learned model, a switching means 212 switches out an old machine-learned model 221 that is stored in a storage unit 22.
机译:在物联网(IoT)的领域中,本发明根据位置切换边缘设备使用的机器学习模型。边缘设备2的发送装置211将由定位单元26生成的位置信息发送到提供装置1。当位置信息从边缘设备2发送到提供装置1时,获取装置111获取位置信息并将其传递给指定装置112。指定装置112指定与该位置信息指示的位置相对应的机器学习模型。供给装置113从模型DB 121读取由指定装置112指定的机器学习模型,并将其提供给将位置信息发送到供给设备1的边缘设备2。当将其提供给边缘设备2时,根据该提供的新的机器学习模型,切换装置212切换出存储在存储单元22中的旧的机器学习模型221。

著录项

  • 公开/公告号WO2019193660A1

    专利类型

  • 公开/公告日2019-10-10

    原文格式PDF

  • 申请/专利权人 UHURU CORPORATION;

    申请/专利号WO2018JP14334

  • 发明设计人 TANIGUCHI MASAHITO;

    申请日2018-04-03

  • 分类号G06N99;

  • 国家 WO

  • 入库时间 2022-08-21 11:52:51

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号