arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1102.3923cs.LGstat.ML
收藏

基于集中度的低秩矩阵重构保证

Concentration-Based Guarantees for Low-Rank Matrix Reconstruction

University of Chicago(芝加哥大学);Toyota Technological Institute at Chicago(芝加哥丰田技术研究院)

Rina Foygel, Nathan Srebro

摘要:

我们考虑的问题,近似重建部分观察,近似低秩矩阵。这个问题最近受到了很多关注,主要是使用跟踪范数作为秩的替代。在这里,我们研究低秩矩阵重建使用的跟踪范数,以及较少研究的最大范数,并根据现有的分析,这些规范的单位球的Rademacher复杂性目前的重建保证。我们展示了这些是如何在几个方面优于最近公布的保证专业分析的基础上。

英文摘要:

We consider the problem of approximately reconstructing a partially-observed, approximately low-rank matrix. This problem has received much attention lately, mostly using the trace-norm as a surrogate to the rank. Here we study low-rank matrix reconstruction using both the trace-norm, as well as the less-studied max-norm, and present reconstruction guarantees based on existing analysis on the Rademacher complexity of the unit balls of these norms. We show how these are superior in several ways to recently published guarantees based on specialized analysis.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑