A new class of Markov random fields enabling lightweight sampling
By: Jean-Baptiste Courbot, Hugo Gangloff, Bruno Colicchio
Potential Business Impact:
Makes computer pictures faster and use less power.
This work addresses the problem of efficient sampling of Markov random fields (MRF). The sampling of Potts or Ising MRF is most often based on Gibbs sampling, and is thus computationally expensive. We consider in this work how to circumvent this bottleneck through a link with Gaussian Markov Random fields. The latter can be sampled in several cost-effective ways, and we introduce a mapping from real-valued GMRF to discrete-valued MRF. The resulting new class of MRF benefits from a few theoretical properties that validate the new model. Numerical results show the drastic performance gain in terms of computational efficiency, as we sample at least 35x faster than Gibbs sampling using at least 37x less energy, all the while exhibiting empirical properties close to classical MRFs.
Similar Papers
Colored Markov Random Fields for Probabilistic Topological Modeling
Machine Learning (Stat)
Models complex connections using colored links.
Particle Monte Carlo methods for Lattice Field Theory
Machine Learning (Stat)
Faster computer simulations for science problems.
DANIEL: A Distributed and Scalable Approach for Global Representation Learning with EHR Applications
Methodology
Lets hospitals share patient data safely.