动态规划模型中高维问题的降维法及疏密格子点法

Reduction Dimension Algorithm and Density Lattice Search Algorithm ——Two Algorithms in Dynamic Programming Modeling While S_k is Higher-dimension

当动态规划模型中的状态变量Sk 为高维时 ,求解时存在“维数灾”。本文介绍的降维法和疏密格子点法 ,采用增加计算机的运行时间来减少对内存的占用 ,可以有效地解决“维数灾”问题。

There is dimension disaster in the solving procedure in dynamic programming modeling while S k is higher dimension. This article introduces two algorithms:reduction dimension algorithm and density lattice search algorithm, which reduce the occupation of memory in computer system by spending more running time. The two algorithms are effictive on solving the problem of dimension disaster.