Information geometry, simulation and complexity in Gaussian random fields

DOI10.1515/MCMA-2016-0107zbMATH Open1339.60057arXiv1703.04464OpenAlexW2403551686MaRDI QIDQ293516FDOQ293516

Authors: Alexandre L. M. Levada 

Publication date: 9 June 2016

Published in: Monte Carlo Methods and Applications (Search for Journal in Brave)

Abstract: Random fields are useful mathematical objects in the characterization of non-deterministic complex systems. A fundamental issue in the evolution of dynamical systems is how intrinsic properties of such structures change in time. In this paper, we propose to quantify how changes in the spatial dependence structure affect the Riemannian metric tensor that equips the model's parametric space. Defining Fisher curves, we measure the variations in each component of the metric tensor when visiting different entropic states of the system. Simulations show that the geometric deformations induced by the metric tensor in case of a decrease in the inverse temperature are not reversible for an increase of the same amount, provided there is significant variation in the system entropy: the process of taking a system from a lower entropy state A to a higher entropy state B and then bringing it back to A, induces a natural intrinsic one-way direction of evolution. In this context, Fisher curves resemble mathematical models of hysteresis in which the natural orientation is pointed by an arrow of time.

Full work available at URL: https://arxiv.org/abs/1703.04464

Recommendations

• scientific article; zbMATH DE number 2016896
• Information geometry and statistical manifold.
• Softening the complexity of entropic motion on curved statistical manifolds
• Information geometric complexity of entropic motion on curved statistical manifolds under different metrizations of probability spaces
• Quantifying networks complexity from information geometry viewpoint

zbMATH Keywords

hysteresisGaussian random fieldsMarkov chain Monte Carlo methodcomplex systemsinformation geometryFisher information

Mathematics Subject Classification ID

Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.) (60J20) Computational methods in Markov chains (60J22) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40) Gaussian processes (60G15) Random fields (60G60) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Simulation of dynamical systems (37M05)

Cites Work

• Statistical mechanics of complex networks
• Differential-geometrical methods in statistics
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Complex networks: structure and dynamics
• Machine learning. A probabilistic perspective
• Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
• The Structure and Function of Complex Networks
• Monte Carlo sampling methods using Markov chains and their applications
• Title not available (Why is that?)
• Equation of state calculations by fast computing machines
• Title not available (Why is that?)
• Exploring complex networks
• Information Theory and Statistical Mechanics
• The exponential statistical manifold: mean parameters, orthogonality and space transformations
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Beitrag zur Theorie des Ferromagnetismus
• Title not available (Why is that?)
• Title not available (Why is that?)
• The geometry of asymptotic inference. With comments and a rejoinder by the author
• Information-theoretic differential geometry of quantum phase transitions
• Science from Fisher Information
• Crystal Statistics. I. A Two-Dimensional Model with an Order-Disorder Transition
• Homogeneous random fields and statistical mechanics
• Phase transition strength through densities of general distributions of zeroes
• On asymptotic normality of pseudo likelihood estimates for pairwise interaction processes
• Random fields and stochastic Lagrangian models. Analysis and applications in turbulence and porous media.
• Fluctuation, dissipation and the arrow of time
• JONES POLYNOMIAL AS A POTTS MODEL PARTITION FUNCTION
• Temporal asymmetry, entropic irreversibility, and finite-time thermodynamics: from Parmenides-Einstein time-reversal symmetry to the Heraclitan entropic arrow of time
• Modeling time's arrow
• Title not available (Why is that?)
• Information geometry on hierarchy of probability distributions
• Stochastic dynamics of complex systems. From glasses to evolution
• Fisher information distance: a geometrical reading
• Random Fields
• Markov Random Fields and Gibbs Ensembles
• Information geometry. Near randomness and near independence
• What are the bounds on space-time non-commutativity?

Cited In (7)

• Monte Carlo Information-Geometric Structures
• Approximate Solutions of Lagrange Multipliers for Information-Theoretic Random Field Models
• Information Geometry of the Gaussian Distribution in View of Stochastic Optimization
• Gaussian random bridges and a geometric model for information equilibrium
• On the Conditional Mutual Information in the Gaussian–Markov Structured Grids
• Information geometric characterization of the complexity of fractional Brownian motions
• Closed-form Bayesian image denoising: improving the adaptive Wiener filter through pairwise Gaussian-Markov random fields

Uses Software

• ElemStatLearn
• PMTK

This page was built for publication: Information geometry, simulation and complexity in Gaussian random fields

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q293516)