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.
这样广义最小二乘法,偏差校正法及伪线性法的离线算法便很容易地以这同一准则导出。
Off-line algorithms of GLS, BCLS and PLR can be derived easily from the same criterion.
应用推荐