Corpus ID: 59604499

Exploiting locality in high-dimensional factorial hidden Markov models

  title={Exploiting locality in high-dimensional factorial hidden Markov models},
  author={Lorenzo Rimella and Nick Whiteley},
  • Lorenzo Rimella, Nick Whiteley
  • Published 2019
  • Computer Science, Mathematics
  • ArXiv
  • We propose algorithms for approximate filtering and smoothing in high-dimensional factorial hidden Markov models. The approximation involves discarding, in a principled way, likelihood factors according a notion of locality in a factor graph associated with the emission distribution. This allows the exponential-in-dimension cost of exact filtering and smoothing to be avoided. We prove that the approximation accuracy, measured in a local total variation norm, is `dimension-free' in the sense… CONTINUE READING
    1 Citations
    Dynamic Bayesian Neural Networks
    • PDF


    Approximate Smoothing and Parameter Estimation in High-Dimensional State-Space Models
    • 2
    • Highly Influential
    Factorial Hidden Markov Models
    • 1,163
    • Highly Influential
    • PDF
    Parallel Subspace Sampling for Particle Filtering in Dynamic Bayesian Networks
    • 16
    • Highly Influential
    • PDF
    Variational Inference: A Review for Statisticians
    • 1,657
    • PDF
    Exploiting the Architecture of Dynamic Systems
    • 57
    • PDF
    Comparison Theorems for Gibbs Measures
    • 17
    • PDF
    Subspace Hierarchical Particle Filter
    • 21
    • PDF
    Asymptotic Normality of Maximum Likelihood and its Variational Approximation for Stochastic Blockmodels
    • 142
    • PDF
    Consistency of maximum-likelihood and variational estimators in the Stochastic Block Model
    • 148
    • PDF