- 专利标题： Discrete feature representation with class priority
-
申请号： US16290250申请日： 2019-03-01
-
公开(公告)号： US11244227B2公开(公告)日： 2022-02-08
- 发明人: Masataro Asai
- 申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION
- 申请人地址： US NY Armonk
- 专利权人： INTERNATIONAL BUSINESS MACHINES CORPORATION
- 当前专利权人： INTERNATIONAL BUSINESS MACHINES CORPORATION
- 当前专利权人地址： US NY Armonk
- 代理机构： Tutunjian & Bitetto, P.C.
- 代理商 Randall Bluestone
- 主分类号： G06N3/08
- IPC分类号： G06N3/08 ; G06K9/62 ; G06N20/00 ; G06N3/04
摘要：
A discrete neural network is trained by training a neural network having an output layer so as to output discrete values. The output layer includes a plurality of nodes. Each node corresponding to one of a plurality of classes. The training includes activating the nodes by priority according to the corresponding class.
公开/授权文献
- US20200279164A1 DISCRETE FEATURE REPRESENTATION WITH CLASS PRIORITY 公开/授权日：2020-09-03
信息查询
