针对BP算法收敛速度慢的特点,在隐含层上加入了关联节点,改善了网络的学习速率和适应能力。
Aiming at the slow convergence rate of BP neural network, append a correlative node on hidden layer, improve the adaptive ability and rate of studying of neural network.
基于输出层函数为线性函数的三层前馈神经网络,结合自适应步长和动量解耦的伪牛顿算法及迭代最小二乘法导出了一种混合算法。
On the basis of both adaptive BP algorithm and Newtons method, Quasi Newton algorithm with adaptive decoupled step and momentum (QNADSM) for feed-forward neural networks is derived.
FCHT利用这些信息计算出触发切换的自适应阀值,通过跨层设计方法使网络层和应用层的切换相衔接。
FCHT compute the self-adapt trigger value using the information. We join the handovers in application layer and in network layer closely by cross-layer method also.
应用推荐