公开(公告)号：US5555439A

公开(公告)日：1996-09-10

申请号：US895928

申请日：1992-06-09

申请人： Junichi Higashino ,  Kazuhiko Kawasaki ,  Miyuki Hara

发明人： Junichi Higashino ,  Kazuhiko Kawasaki ,  Miyuki Hara

IPC分类号： G06F15/18 ,  G06G7/60 ,  G06N3/08 ,  G06N99/00

CPC分类号： G06N3/084

摘要： A neural network learning system using back propagation and a method of learning pattern showing are disclosed. In the case where a supervised signal contains an error or a pattern difficult to learn, the error is detected and the particular pattern is automatically removed during calculations of learning iterations to conduct rightly and accelerate the learning. The learning history for each pattern is stored to detect inconsistent and difficult-to-learn patterns, which are prevented from being shown to the network by a pattern showing control during the next learning iteration. As a result, an inconsistent or difficult-to-learn pattern which may be contained in a learning pattern set of input and supervised patterns is removed during learning iterations thereby to permit early completion of the learning process.

摘要翻译： 公开了一种使用反向传播的神经网络学习系统和学习模式显示方法。 在监督信号包含错误或难以学习的模式的情况下，检测到错误，并且在计算学习迭代期间自动去除特定模式以正确地行进并加速学习。 存储每个模式的学习历史以检测不一致和难以学习的模式，这些模式通过在下一次学习迭代期间显示控制的模式被阻止显示给网络。 因此，在学习迭代期间，消除了可能包含在输入和监督模式的学习模式集合中的不一致或难以学习的模式，从而允许学习过程的早期完成。