论文部分内容阅读
有关极大熵聚类算法收敛性的研究是理论研究的一个热点问题,有的学者认为迭代序列的极限点有可能不是目标函数的严格局部极值点.针对这个问题,文中用科学计算软件对相关文献中给出的例子进行了实验,说明该例子并不能否定极大熵聚类算法收敛性定理.最后,从理论上给出了极大熵聚类算法收敛性定理的一个证明.
The research on the convergence of the maximum entropy clustering algorithm is a hot issue in theoretical research, and some scholars think that the limit point of the iterative sequence may not be the strict local extremum point of the objective function.In this paper, we use the scientific computing software Experiments have been done on the examples given in the relevant literature, which shows that the example does not negate the convergence theorem of the maximum entropy clustering algorithm. Finally, a theorem is given to prove the convergence theorem of the maximum entropy clustering algorithm.