Spectral Properties of Large Random Matrices

大型随机矩阵的谱特性

• 批准号: 0707145
• 负责人: Alexander Soshnikov
• 金额: $ 14万
• 依托单位: University of California-Davis
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2007
• 资助国家: 美国
• 起止时间: 2007-08-01 至 2010-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0707145&HistoricalAwards=false
• 关键词: Spectral; Properties; Large; Random; Matrices

The principal investigator works on problems in Random Matrix Theory and Probability Theory with the main emphasis on statistical properties of the eigenvalues of large random matrices. The P.I. intends to continue his studies of the distribution of the largest eigenvalues in Wigner, sample covariance, and related ensembles of random matrices. The proposed research is closely related to the Tracy-Widom distribution of the largest eigenvalues of Hermitian and real symmetric random matrices that appears in many important problems in physics, statistics, and theoretical computer science: crystal shapes, exclusion processes, directed polimers in random media, principal component analysis. In particular, the P.I. proposes to establish Tracy-Widom distribution in some important ensembles of random matrices. Recently, the P.I. discovered Poisson statistics of the largest eigenvalues in random matrices with power-law tailed elements. Remarkably, such matrices very often appear in financial applications. The P.I. intends to continue his studies of heavy-tailed random matrices and, in particular, to study in detail the transition form the Tracy-Widom regime to the Poisson regime. The random matrix models studied in the project come from or have applications in nuclear physics (statistics of energy levels of heavy nuclei), mathematical statistics (principal component analysis), theoretical computer science (computational complexity, statistical analysis of errors, linear numerical algorithms), population biology, mathematical finance, and solid state physics (modeling transport properties of small metallic particles and quantum dots). The importance of the field has increased significantly in the last ten years as many different areas of mathematics and physics including combinatorics, representation theory, number theory, quantum gravity, integrable systems, and random growth models have been shown to possess deep and fruitful connections to random matrix theory.

主要研究随机矩阵理论和概率论中的问题，主要侧重于大型随机矩阵特征值的统计特性。 私家侦探打算继续他的研究分布的最大特征值的维格纳，样本协方差，以及相关的合奏随机矩阵。建议的研究是密切相关的Tracy-Widom分布的最大特征值的厄米特和真实的对称随机矩阵，出现在许多重要的问题，在物理学，统计学和理论计算机科学：晶体形状，排斥过程，定向聚合物在随机介质中，主成分分析。 特别是，P.I.提出在一些重要的随机矩阵集合中建立Tracy-Widom分布。 最近，PI发现了具有幂律尾元素的随机矩阵的最大特征值的泊松统计。值得注意的是，这种矩阵经常出现在金融应用中。 私家侦探打算继续他的研究重尾随机矩阵，特别是，研究详细的过渡形式的特雷西Widom制度的泊松制度。该项目研究的随机矩阵模型来自或应用于核物理（重核能级统计），数理统计（主成分分析），理论计算机科学（计算复杂性，误差统计分析，线性数值算法），人口生物学，数学金融学和固态物理学（模拟小金属粒子和量子点的传输特性）。 该领域的重要性在过去十年中显着增加，因为数学和物理学的许多不同领域，包括组合学，表示论，数论，量子引力，可积系统和随机增长模型已被证明与随机矩阵理论有着深刻而富有成效的联系。