arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.05628cs.LGcs.CG
收藏

置信度最优的随机嵌入

Confidence-Optimal Random Embeddings

University of Luxembourg(卢森堡大学)

Maciej Skorski

摘要:

Johnson和Lindenstrauss关于随机嵌入的开创性成果在应用和理论计算机科学中得到了深入的研究。尽管有大量的文献,我们仍然缺乏对随机预测的统计特性的完整理解;一个特别有趣的问题是:为什么理论界远远落后于经验观察到的性能?出于这个问题的动机,这项工作开发了具有最佳,数据无关,统计置信区间的约翰逊-林登施特劳斯分布。对于任何给定的数据维数、嵌入维数和失真容限,这些边界在数值上是最好的。它们在统计准确性方面改进了先前的工作,并准确地确定了数据遗忘方法的不适用范围。此外,相应的投影矩阵是有效的采样。构造依赖于正交矩阵,证明使用了单位球的某些优雅性质。在这项工作中介绍的以下技术是独立的兴趣:a)一个紧凑的表达失真的奇异特征值的投影矩阵,b)连接单位球和狄利克雷分布和c)反浓度边界的狄利克雷分布的参数化。除了技术贡献外,本文还介绍了应用程序和数值计算以及Python中的工作实现。

英文摘要:

The seminal result of Johnson and Lindenstrauss on random embeddings has been intensively studied in applied and theoretical computer science. Despite that vast body of literature, we still lack of complete understanding of statistical properties of random projections; a particularly intriguing question is: why are the theoretical bounds that far behind the empirically observed performance? Motivated by this question, this work develops Johnson-Lindenstrauss distributions with optimal, data-oblivious, statistical confidence bounds. These bounds are numerically best possible, for any given data dimension, embedding dimension, and distortion tolerance. They improve upon prior works in terms of statistical accuracy, as well as exactly determine the no-go regimes for data-oblivious approaches. Furthermore, the corresponding projection matrices are efficiently samplable. The construction relies on orthogonal matrices, and the proof uses certain elegant properties of the unit sphere. The following techniques introduced in this work are of independent interest: a) a compact expression for distortion in terms of singular eigenvalues of the projection matrix, b) a parametrization linking the unit sphere and the Dirichlet distribution and c) anti-concentration bounds for the Dirichlet distribution. Besides the technical contribution, the paper presents applications and numerical evaluation along with working implementation in Python.

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。

↑