Multiscale sparse microcanonical models

  • Joan Bruna

    New York University, USA
  • Stéphane Mallat

    Collège de France and École Normale Supérieure, Paris, France


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.

Cite this article

Joan Bruna, Stéphane Mallat, Multiscale sparse microcanonical models. Math. Stat. Learn. 1 (2018), no. 3/4, pp. 257–315

DOI 10.4171/MSL/7