课题基金 / 基金详情

Random Hessians and Jacobians: theory and applications

Random Hessians and Jacobians: theory and applications
随机 Hessians 和 Jacobian:理论与应用
批准号:
EP/V002473/1
负责人:
Yan Fyodorov
金额:
$105.08万
依托单位:
依托单位国家:
英国
项目类别:
Research Grant
财政年份:
2021
资助国家:
英国
项目状态:
未结题
起止时间:
2021 至 --

项目摘要

项目成果

Yan Fyodorov的其他基金

相关文献

中文摘要
翻译
复杂“景观”的性质,即定义在非常高维空间上的随机函数,最近引起了相当大的关注,例如在深度机器学习和优化理论中。特别是,人们可能会对给定“高度”的“谷”(即局部最小值),将它们分开的“脊”或障碍,以及更一般的“临界点”(鞍和最大值)的数量感兴趣。在描述地形的几何特征方面,尤其是在接近临界点的地方,一个重要的角色是由二阶导数矩阵扮演的,即黑森矩阵。例如,它决定了这些景观中的梯度下降动态,这对搜索算法有许多实际应用。根据环境的不同,景观可以对应于物理系统的能量、机器学习算法的损失函数、优化问题的成本函数或生物系统的适应度函数。在临界点分析中,海森特征多项式的模数自然出现。同样地,要描述复杂动力系统(例如许多相互作用物种的群落)中的平衡,需要研究更一般的,不对称的,雅可比矩阵的性质,其中Hessians只是一个特例。雅可比矩阵与系统在小扰动下的稳定性问题密切相关,因此是非常基本的。注意,与光谱实数且特征向量构成正交集的Hessians相比,雅可比矩阵通常具有复特征值和左右特征向量的双正交集。随机环境下相关的“特征向量非正交性”的研究不仅与复杂系统的稳定性有关,也与混沌波散射和随机激光有关。目前的研究计划主要集中在分析随机矩阵和算子的各种性质,主要是由随机景观的Hessians或各种来源的随机雅可比矩阵引起的。
英文摘要
Properties of complicated 'landscapes', i.e. randomfunctions defined on very high dimensional spaces, have recently attracted considerable attention, e.g. in theory of Deep Machine Learning and Optimization. In particular, one may be interested in number of 'valleys' (i.e. local minima) at a given 'height', 'ridges' or barriers separating them, and more generally 'critical points' (saddles and maxima). An important role in characterising geometry of the landscapes, especially close to the critical points, is played by the matrix of second derivatives known as the Hessian. It determines e.g. the gradient descent dynamics within these landscapes, which has many practical applications for search algorithms. Depending on the context, the landscape can correspond to the energy of a physical system, to the loss function of a machine-learning algorithm, to the cost function of an optimization problem, or tothe fitness function of a biological system. In the analysis of critical points the (modulus of) the characteristic polynomial of the Hessian appears naturally. Similarly, to characterize equilibria in complicated dynamical systems (e.g. communities of many interacting species) requires investigating properties of more general, asymmetric, Jacobian matrices, for which Hessians are only a special case. Jacobians are deeply related to questions of stability of systems under small perturbations, and as such are very fundamental. Note that in contrast to Hessians whose spectra are real and eigenvectors form an orthogonal set, the Jacobians have in general complex eigenvalues and bi-orthogonal set of left and right eigenvectors. The studies of the associated 'eigenvector non-orthogonality' in random setting turn out to be relevant both for complex systems stability as well as to chaotic wave scattering and random lasing. The present research proposal is mainly centred around analysis of various properties of random matrices and operators, mostly arising via Hessians of random landscapes, or random Jacobians of various origin.
期刊论文(10)
专著(0)
科研奖励(0)
会议论文
Statistics of Complex Wigner Time Delays as a counter of S-matrix poles: Theory and Experiment
作为 S 矩阵极点计数器的复杂维格纳时滞统计:理论与实验
DOI: 10.48550/arxiv.2106.15469
发表时间: 2021
期刊:
影响因子: --
作者: [Chen L]
通讯作者: Chen L
Universal microscopic spectrum of the unquenched QCD Dirac operator at finite temperature
有限温度下未淬火 QCD 狄拉克算子的通用显微光谱
DOI: 10.1007/jhep12(2021)128
发表时间: 2021
期刊: Journal of High Energy Physics
影响因子: 5.4
作者: [Akemann G]
通讯作者: Akemann G
Resonances in a single-lead reflection from a disordered medium: s -model approach
无序介质中单导联反射的共振:s 模型方法
DOI: 10.1016/j.aop.2023.169568
发表时间: 2023
期刊: Annals of Physics
影响因子: 3
作者: [Fyodorov Y]
通讯作者: Fyodorov Y
On the Density of Complex Eigenvalues of Wigner Reaction Matrix in a Disordered or Chaotic System with Absorption
关于有吸收的无序或混沌系统中维格纳反应矩阵的复特征值密度
DOI: 10.12693/aphyspola.144.447
发表时间: 2023
期刊: Acta Physica Polonica A
影响因子: 0.7
作者: [Fyodorov AFFSTART Y]
通讯作者: Fyodorov AFFSTART Y
The Many Faces of Random Characteristic Polynomials
  • 批准号:
    EP/N009436/1
  • 项目类别:
    Research Grant
  • 资助金额:
    $60.89万
  • 财政年份:
    2016
  • 负责人:
    Yan Fyodorov
  • 依托单位:
Insights into Disordered Landscapes via Random Matrix Theory and Statistical Mechanics
  • 批准号:
    EP/J002763/1
  • 项目类别:
    Research Grant
  • 资助金额:
    $46.64万
  • 财政年份:
    2012
  • 负责人:
    Yan Fyodorov
  • 依托单位:
A proposal for the visit of Dr. Vladimir Al. Osipov: From Random Matrices to Random Landscapes
  • 批准号:
    EP/G022496/1
  • 项目类别:
    Research Grant
  • 资助金额:
    $1.81万
  • 财政年份:
    2009
  • 负责人:
    Yan Fyodorov
  • 依托单位: