SOFIM: Stochastic Optimization Using Regularized Fisher Information Matrix [PDF]
2024 International Joint Conference on Neural Networks (IJCNN)This paper introduces a new stochastic optimization method based on the regularized Fisher information matrix (FIM), named SOFIM, which can efficiently utilize the FIM to approximate the Hessian matrix for finding Newton's gradient update in large-scale stochastic optimization of machine learning models.
C. Gayathri +5 more
openalex +3 more sources
Fisher Information and Logarithmic Sobolev Inequality for Matrix Valued Functions [PDF]
Annales Henri Poincaré, 2018 We prove a version of Talagrand's concentration inequality for subordinated sub-Laplacian on a compact Riemannian manifold using tools from noncommutative geometry. As an application, motivated by quantum information theory, we show that on a finite dimensional matrix algebra the set of self-adjoint generators satisfying a tensor stable modified ...
Li Gao, Marius Junge, Nicolas LaRacuente
openalex +4 more sources
On weighted Fisher information matrix properties [PDF]
, 2016 In this paper, we review Fisher information matrices properties in weighted version and discuss inequalities/bounds on it by using reduced weight functions. In particular, an extended form of the Fisher information inequality previously established in [6] is given.
Mark Kelbert +2 more
openalex +3 more sources
Comparing the Fisher information matrix in record values and random observations for the general class of exponentiated distributions [PDF]
Journal of Statistical Theory and Applications (JSTA), 2017 In this paper, the Fisher information matrix (FIM) contained in n record values is considered for the two parameter distributions belong to the exponentiated and inverse exponentiated class of distributions. The problem of existence and uniqueness of the
Fatih Kızılaslan
doaj +3 more sources
Physical Review ResearchA group of techniques known by the general name of single-molecule localization microscopy reaches a nanometer-scale spatial resolution of point light emitters, well below the diffraction limit of traditional microscopy.
Dmitri B. Horoshko +3 more
doaj +2 more sources
Asymptotic Fisher information matrix of Markov switching VARMA models
Journal of Multivariate Analysis, 2017 zbMATH Open Web Interface contents unavailable due to conflicting licenses.
Maddalena Cavicchioli
openalex +4 more sources
Quantifying entanglement dimensionality from the quantum Fisher information matrix [PDF]
Entanglement is known to be an essential resource for a number of tasks, including quantum-enhanced metrology, and can thus be quantified by figures of merit related to those tasks. In quantum metrology this is emphasized by the connections between the quantum Fisher information (QFI), providing the ultimate bounds of precision, and multipartite ...
Shuya Du +4 more
openalex +3 more sources
The effect of Fisher information matrix approximation methods in population optimal design calculations. [PDF]
J Pharmacokinet Pharmacodyn, 2016 Strömberg EA, Nyberg J, Hooker AC.
europepmc +3 more sources
UAV Path Optimization for Angle-Only Self-Localization and Target Tracking Based on the Bayesian Fisher Information Matrix. [PDF]
Sensors (Basel)Dogancay K, Hmam H.
europepmc +3 more sources
An Algorithm for the Fisher Information Matrix of a VARMAX Process
Algorithms, 2023 In this paper, an algorithm for Mathematica is proposed for the computation of the asymptotic Fisher information matrix for a multivariate time series, more precisely for a controlled vector autoregressive moving average stationary process, or VARMAX ...
André Klein, Guy Mélard
doaj +1 more source