The pheromone - based parameterized probabilistic model for the aco algorithm is presented as the solution construction graph that the combinatorial optimization problem can be mapped on . based on the solution construction graph , the unified framework of the aco algorithm is presented . an iterative update procedure of the solutions distribution in the problem ' s probabilistic model is proposed , that will converge to the optimal solutions with probability one , then the minimum cross - entropy pheromone update rule is proposed to approximate the iterative update procedure by minimizing the cross - entropy distance and monte - carlo sampling 基于解空間參數(shù)化概率分布模型,首先提出了一個以概率1收斂于最優(yōu)解的解空間概率分布的迭代更新過程,然后提出了通過最小化不同分布間的交互熵距離以及蒙特卡洛采樣來逼近此迭代過程的最小交互熵信息素更新規(guī)則,接著分別給出了弧模式以及結(jié)點模式信息素分布模型下的最小交互熵等式。
The cores of the task planning in joint campaign are tasks and actions . the aim of the task planning in joint campaign is how to distribute the combat resources to each task efficiently . taking every constraints of the distribution process into account , this distribution problem is actually a combinatorial optimization problem 聯(lián)合作戰(zhàn)任務(wù)計劃的核心是任務(wù)和行動,其目的就是如何有效的將作戰(zhàn)資源分配到各個任務(wù)中去,考慮到分配過程中的各種約束,對這一問題的求解即是解決一個組合優(yōu)化問題。