期刊文献+

线性约束极大熵问题的正则约束凝聚函数法

The Regularized Constraint Aggregation Method for Maximum Entropy Subject to Linear Constraints
下载PDF
导出
摘要 研究了求解具有线性约束极大熵问题的约束凝聚函数法的正则化技术。根据在近似点算法中使用约束凝聚函数法的思想,提出了求解具有线性约束极大熵问题的正则约束凝聚函数法。对具有线性约束极大熵问题的等价问题建立了正则约束凝聚函数。分别在没有正则条件下和一些正则条件下,给出了其收敛性分析。 In this paper,we study regularization techniques for the constraint aggregation method for solving maximum entropy problem subject to linear constraints.We use the constraint aggregation principle in the general framework of the proximal point method and propose the regularized constraint aggregation method for solving maximum entropy problem subject to linear constraints.At first,the regularized constraint aggregation function is established for the equlivant problem of maximum entropy problem subject to l...
出处 《上海电机学院学报》 2008年第3期227-230,共4页 Journal of Shanghai Dianji University
基金 国家自然科学基金重点资助项目(50745039)
关键词 极大熵 正则化 近似点方法 约束凝聚函数 非光滑最优化 maximum entropy regularization proximal-point method constraint aggregation nonsmooth optimization
  • 相关文献

参考文献8

  • 1[1]Gull S F,Skilling J.Maximum Entropy Method in Image Processing[J].IEEE Proceeding,1984,131:646-659.
  • 2[2]Jaynes E T.On The Rationale of Maximum Entropy-Methods[J].IEEE Proceedings,1987,70(9):939-952.
  • 3[3]Yang H,Wong S C.Estimation of the Most Likely Equilibrium Traffic Queueing Pattern in a Capacity-constrained Network[J].Annals of Operations Research,1999,87:73-85.
  • 4[4]Erikssion J.A Note on Solution of Large Sparse Maximum Entropy Problems with Linear Equality Constraints[J].Mathematical Programming,1980,18(1):146-154.
  • 5[5]Powell M J D.An Algorithm for Maximizing Entropy Subject to Simple Bounds[J].Mathematical Programming,1988,42(1-3):171-180.
  • 6焦建六.求解一类极大熵问题的一种新算法[J].系统工程,2000,18(3):69-72. 被引量:5
  • 7[7]Bazarra M S,Sherali H D,Shetty C M.Nonlinear Programming:Theory and Algorithms[M].New York:Wiley,1993.
  • 8[8]Wasan M T.Stochastic Approximation[M].London:Cambridge Univ.Press,1969.

二级参考文献2

共引文献4

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部