发明授权
US5644681A Learning method for neural network having discrete interconnection
strengths
失效
具有离散互连强度的神经网络的学习方法
- 专利标题： Learning method for neural network having discrete interconnection strengths
- 专利标题（中）： 具有离散互连强度的神经网络的学习方法
-
申请号： US363597申请日： 1994-12-23
-
公开(公告)号： US5644681A公开(公告)日： 1997-07-01
- 发明人: Masanobu Takahashi , Shuichi Tai , Keisuke Kojima , Toshio Shinnishi , Kazuo Kyuma
- 申请人： Masanobu Takahashi , Shuichi Tai , Keisuke Kojima , Toshio Shinnishi , Kazuo Kyuma
- 申请人地址： JPX Tokyo
- 专利权人： Mitsubishi Denki Kabushiki Kaisha
- 当前专利权人： Mitsubishi Denki Kabushiki Kaisha
- 当前专利权人地址： JPX Tokyo
- 优先权： JPX1-327704 19891218
- 主分类号： G06N3/08
- IPC分类号： G06N3/08 ; G06F15/18
摘要：
A learning method for a neural network, in which at least a portion of the interconnection strength between neurons takes discrete values, includes the steps of updating an imaginary interconnection strength taking continuous values by using the quantity of update of the interconnection strength which has been calculated by using the discrete interconnection strength, and discretizing the imaginary interconnection strength so as to provide a novel interconnection strength.
信息查询
