期刊文献+

The bounds of restricted isometry constants for low rank matrices recovery 被引量:6

The bounds of restricted isometry constants for low rank matrices recovery
原文传递
导出
摘要 This paper discusses conditions under which the solution of linear system with minimal Schatten-p norm, 0 〈 p ≤ 1, is also the lowest-rank solution of this linear system. To study this problem, an important tool is the restricted isometry constant (RIC). Some papers provided the upper bounds of RIC to guarantee that the nuclear-norm minimization stably recovers a low-rank matrix. For example, Fazel improved the upper bounds to δ4Ar 〈 0.558 and δ3rA 〈 0.4721, respectively. Recently, the upper bounds of RIC can be improved to δ2rA 〈 0.307. In fact, by using some methods, the upper bounds of RIC can be improved to δ2tA 〈 0.4931 and δrA 〈 0.309. In this paper, we focus on the lower bounds of RIC, we show that there exists linear maps A with δ2rA 〉1√2 or δrA 〉 1/3 for which nuclear norm recovery fail on some matrix with rank at most r. These results indicate that there is only a little limited room for improving the upper bounds for δ2rA and δrA.Furthermore, we also discuss the upper bound of restricted isometry constant associated with linear maps A for Schatten p (0 〈 p 〈 1) quasi norm minimization problem. This paper discusses conditions under which the solution of linear system with minimal Schatten-p norm, 0 < p 1, is also the lowest-rank solution of this linear system. To study this problem, an important tool is the restricted isometry constant (RIC). Some papers provided the upper bounds of RIC to guarantee that the nuclear-norm minimization stably recovers a low-rank matrix. For example, Fazel improved the upper bounds to δA4r < 0.558 and δA3r < 0.4721, respectively. Recently, the upper bounds of RIC can be improved to δA2r < 0.307. In fact, by using some methods, the upper bounds of RIC can be improved to δA2r < 0.4931 and δ A r < 0.309. In this paper, we focus on the lower bounds of RIC, we show that there exists linear maps A with δA2r > 1/ 2^(1/2) or δAr > 1/3 for which nuclear norm recovery fail on some matrix with rank at most r. These results indicate that there is only a little limited room for improving the upper bounds for δA2r and δAr . Furthermore, we also discuss the upper bound of restricted isometry constant associated with linear maps A for Schatten p (0 < p < 1) quasi norm minimization problem.
出处 《Science China Mathematics》 SCIE 2013年第6期1117-1127,共11页 中国科学:数学(英文版)
基金 supported by National Natural Science Foundation of China (Grant Nos.91130009, 11171299 and 11041005) National Natural Science Foundation of Zhejiang Province in China (Grant Nos. Y6090091 and Y6090641)
关键词 restricted isometry constants low-rank matrix recovery Schatten-p norm nuclear norm com-pressed sensing convex optimization 矩阵 常数 等距 Schatten 最小化问题 回收 线性系统 线性映射
  • 相关文献

参考文献30

  • 1Argyriou A, Micchelli C, Pontil M. Convex multi-task feature learning. Machine Learning, 2008, 73:243-272.
  • 2Beck C, Andrea R. Computational study and comparisons of LFT reducibility methods. In: Proceedings of the American Control Conference. Michigan: American Automatic Control Council, 1998, 1013- 1017.
  • 3Cai J F, Cand's E J, Shen Z W. A singular value thresholding algorithm for matrix completion. SIAM J Optim, 2010, 20:1956-1982.
  • 4Cai T T, Wang L, Xu G W. Shifting inequality and recovery of sparse signals. IEEE Trans Signal Process, 2010, 58: 1300-1308.
  • 5Cai T T, Wang L, Xu G W. New bounds for restricted isometry constants. IEEE Trans Inform Theory, 2010, 56: 4388-4394.
  • 6Cand's E J. Compressive sampling. In: Proceedings of International Congress of Mathematicians, vol. 3. Madrid: European Mathematical Society Publishing House, 2006, 1433-1452.
  • 7Cand's E J. The restricted isometry property and its implications for compressed sensing. C R Acad Sci Paris Ser 1, 2008, 346:589 592.
  • 8Cand's E J, Plan Y. Tight oracle bounds for low-rank recovery from a minimal number of random measurements. IEEE Trans Inform Theory, 2009, 57:2342- 2359.
  • 9Cand's E J, Recht B. Exact matrix completion via convex optimization. Found Comput Math, 2008, 9:717-772.
  • 10Cand's E J, Tao T. Decoding by linear programming. IEEE Trans Inform Theory, 2005, 51:4203 -4215.

同被引文献23

引证文献6

二级引证文献5

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