4.5 Article

The -Truncated -Moment Problem

期刊

FOUNDATIONS OF COMPUTATIONAL MATHEMATICS
卷 14, 期 6, 页码 1243-1276

出版社

SPRINGER
DOI: 10.1007/s10208-014-9225-9

关键词

A-truncated multisequence; Flat extension; Localizing matrix; Semidefinite program; Completely positive matrix; Sums of even powers

资金

  1. NSF [DMS-0844775, DMS-1417985]
  2. Division Of Mathematical Sciences
  3. Direct For Mathematical & Physical Scien [0844775] Funding Source: National Science Foundation
  4. Division Of Mathematical Sciences
  5. Direct For Mathematical & Physical Scien [1417985] Funding Source: National Science Foundation

向作者/读者索取更多资源

Let be a finite set, and be a compact semialgebraic set. An -truncated multisequence (-tms) is a vector indexed by elements in . The -truncated -moment problem (-TKMP) concerns whether or not a given -tms admits a -measure , i.e., is a nonnegative Borel measure supported in such that for all . This paper proposes a numerical algorithm for solving -TKMPs. It aims at finding a flat extension of by solving a hierarchy of semidefinite relaxations for a moment optimization problem, whose objective is generated in a certain randomized way. If admits no -measures and is -full (there exists that is positive on ), then is infeasible for all big enough, which gives a certificate for the nonexistence of representing measures. If admits a -measure, then for almost all generated , this algorithm has the following properties: i) we can asymptotically get a flat extension of by solving the hierarchy ; ii) under a general condition that is almost sufficient and necessary, we can get a flat extension of by solving for some ; iii) the obtained flat extensions admit a -atomic -measure with . The decomposition problems for completely positive matrices and sums of even powers of real linear forms, and the standard truncated -moment problems, are special cases of -TKMPs. They can be solved numerically by this algorithm.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.5
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据