算法在不收敛情况下以误差最小解为计算结果,提高了DTS的鲁棒性。
To enhance the robustness of DTS, the solution with the minimum error is taken as the result when convergence cannot be found.
方法应用单调迭代技术结合上下解方法讨论最大解与最小解的存在性。
Methods The method of upper and lower solutions and the monotone iterative technique were used to establish our results.
从大自然植物生长中得到启发,提出了一种求解非线性整数规划全局最小解的仿生算法。
Be inspired from natural plant growth, this paper proposes a bionic algorithm for solving the global optimization of nonlinear integer programming.
本文讨论了在n维非对称的质量损失函数下,调整参数设计的可行性,求出了质量损失的最小解。
In the paper, Asymmetric quadric loss function is put forward under Normal assumption, adjustment parameter is defined, and its best value is calculated which minimizes the total quality lose.
研究一类四阶微分方程解的存在性,利用上下解及单调迭代的方法,得出这类四阶方程的最大解和最小解的存在。
We obtain the existence of extremal solutions of the boundary value problem by using the method of lower and upper solutions coupled with monotone iterative technique.
神经网络参数估计有许多训练算法,BP算法是前向多层神经网络的典型算法,但BP算法有时会陷入局部最小解。
There are a few training algorithms for parameter estimation of neural networks, in which Back Propagation(BP)algorithm is the typical algorithm for feed-forward multi-layer neural networks.
利用该方法良好的收敛性,可以借助于电子计算机快速的构造目前已知的具有最大码率和最小解码时延的多天线复正交空时分组编码矩阵。
By the convergence of the method, we can fast construct a COSTBC matrix with the maximal rates and the minimal decoding delays that we all know now recur to the computer.
利用该方法良好的收敛性,可以借助于电子计算机快速的构造目前已知的具有最大码率和最小解码时延的多天线复正交空时分组编码矩阵。
By the convergence of the method, we can fast construct a COSTBC matrix with the maximal rates and the minimal decoding delays that we all know now recur to the computer.
应用推荐