A subspace truncated-Newton algorithm for large-scale bound constrained optimization is proposed.
给出了大规模界约束优化的一个子空间截断牛顿法。
The results showed that the practically mixed algorithm was better than simple GA or Gauss-Newton algorithm.
结果表明合成算法优于单纯的遗传算法或高斯-牛顿法，在实践中有一定的应用价值。
The disadvantages of quasi-Newton algorithm is the great memory, so for large problems, memory difficulties may be encountered.
拟牛顿算法的缺点是所需存储量较大，对于大型问题，可能遇到存储方面的困难。
The EML registration is achieved by two step recursive optimization. The quick convergence is assured through the improved Gauss Newton algorithm.
该算法通过两步递 归最优化方法来实现 ，并采用改进的高斯—牛顿法来确保算法的快速收敛 性。
By Newton algorithm and the analysis for asymptotic solution, this paper presents a direct formula for calculating critical depths of trapezoidal open channels.
依据牛顿算法及渐近解分析，本文提出了一种梯形明渠临界流方程的直接算法。
By Newton algorithm and the analysis for asymptotic solution, this paper presents a direct formula for calculating critical depths of trapezoidal open channels.
依据牛顿算法及渐近解分析，本文提出了一种梯形明渠临界流方程的直接算法。
应用推荐