BP神经网络，改进后的高等算法它的学习规则是使用最速下降法，通过反向传播来不断调整网络的权值和阈值，使网络的误差平方和最小。- BP neural network algorithm improved its higher learning rule is to use the steepest descent method, through a reverse spread to constantly adjust the network weights and thresholds, s

mpi多线程运用最速下降法求共轭梯度，16个线程，最多循环1000次。-mpi multi-threaded application steepest descent method the conjugate gradient, 16 threads, up to 1000 cycles.

通过最速下降算法研究基于梯度的自适应方法,最速下降是递归的,(Through the study of steepest descent algorithm based on gradient adaptive method, the steepest decline is recursive,)