Multiscale sparse microcanonical models
From MaRDI portal
Density estimation (62G07) Neural nets and related approaches to inference from stochastic processes (62M45) Higher-dimensional and -codimensional surfaces in Euclidean and related (n)-spaces (53A07) Lattice systems (Ising, dimer, Potts, etc.) and systems on graphs arising in equilibrium statistical mechanics (82B20) Renormalization group methods in equilibrium statistical mechanics (82B28)
Abstract: We study approximations of non-Gaussian stationary processes having long range correlations with microcanonical models. These models are conditioned by the empirical value of an energy vector, evaluated on a single realization. Asymptotic properties of maximum entropy microcanonical and macrocanonical processes and their convergence to Gibbs measures are reviewed. We show that the Jacobian of the energy vector controls the entropy rate of microcanonical processes. Sampling maximum entropy processes through MCMC algorithms require too many operations when the number of constraints is large. We define microcanonical gradient descent processes by transporting a maximum entropy measure with a gradient descent algorithm which enforces the energy conditions. Convergence and symmetries are analyzed. Approximations of non-Gaussian processes with long range interactions are defined with multiscale energy vectors computed with wavelet and scattering transforms. Sparsity properties are captured with norms. Approximations of Gaussian, Ising and point processes are studied, as well as image and audio texture synthesis.
Recommendations
- Interactions in a multi-scale representation of sparse media: from mechanics to thermodynamics
- Microcanonical solution of lattice models with long-range interactions
- Resolving the Multitude of Microscale Interactions Accurately Models Stochastic Partial Differential Equations
- Microcanonical finite-size scaling
- scientific article; zbMATH DE number 822323
- Coarse-grained stochastic processes for microscopic lattice systems
- Finite dimensional models for random microstructures
- scientific article; zbMATH DE number 1790432
- Conservative tightly-coupled simulations of stochastic multiscale systems
- Mesoscopic scales in hierarchical configuration models
Cites work
- scientific article; zbMATH DE number 410740 (Why is no real title available?)
- scientific article; zbMATH DE number 3903723 (Why is no real title available?)
- scientific article; zbMATH DE number 4074075 (Why is no real title available?)
- scientific article; zbMATH DE number 19378 (Why is no real title available?)
- A Robust Gradient Sampling Algorithm for Nonsmooth, Nonconvex Optimization
- A dozen de Finetti-style results in search of a theory
- A note about the uniform distribution on the intersection of a simplex and a sphere
- A parametric texture model based on joint statistics of complex wavelet coefficients
- A probabilistic approach to the geometry of the \(\ell^n_p\)-ball
- Convergence of the Iterates of Descent Methods for Analytic Cost Functions
- Crystal Statistics. I. A Two-Dimensional Model with an Order-Disorder Transition
- Deep Scattering Spectrum
- From Newton to Boltzmann: hard spheres and short-range potentials
- Gibbs measures and phase transitions.
- Global convergence of ADMM in nonconvex nonsmooth optimization
- Gradient descent only converges to minimizers: non-isolated critical points and invariant regions
- Graphical models, exponential families, and variational inference
- Group invariant scattering
- Hamburger moment problem and maximum entropy: on the existence conditions
- Information Theory and Statistical Mechanics
- Intermittent process analysis with scattering moments
- Introduction to the functional renormalization group
- Markov chains and mixing times. With a chapter on ``Coupling from the past by James G. Propp and David B. Wilson.
- Microcanonical distributions for lattice gases
- Phase harmonic correlations and convolutional neural networks
- Power spectra of random spike fields and related processes
- Random Phase Textures: Theory and Synthesis
- Sard theorems for Lipschitz functions and applications in optimization
- Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images
- Time evolution of large classical systems
- Wavelets and Renormalization
Cited in
(13)- Deep learning architectures for nonlinear operator functions and nonlinear inverse problems
- Probabilistic symmetries and invariant neural networks
- Wavelet phase harmonics
- Patch Redundancy in Images: A Statistical Testing Framework and Some Applications
- Maximum entropy models from phase harmonic covariances
- Particle gradient descent model for point process generation
- Complete set of translation invariant measurements with Lipschitz bounds
- Path shadowing Monte Carlo
- Scale dependencies and self-similar models with wavelet scattering spectra
- Central and Noncentral Limit Theorems Arising from the Scattering Transform and Its Neural Activation Generalization
- Maximum entropy methods for texture synthesis: theory and practice
- Stable separation of orbits for finite abelian group actions
- Geometric scattering on measure spaces
This page was built for publication: Multiscale sparse microcanonical models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2319816)
