Colored Markov Random Fields for Probabilistic Topological Modeling
This addresses the problem of modeling complex systems with topological dependencies for researchers in probabilistic modeling and topological signal processing, representing an incremental extension of Gaussian Markov Random Fields.
The paper tackled the limitation of canonical Probabilistic Graphical Models in handling variables on topological spaces by introducing Colored Markov Random Fields (CMRFs), which model both conditional and marginal dependencies among Gaussian edge variables, and demonstrated benefits in a distributed estimation case study over a physical network.
Probabilistic Graphical Models (PGMs) encode conditional dependencies among random variables using a graph -nodes for variables, links for dependencies- and factorize the joint distribution into lower-dimensional components. This makes PGMs well-suited for analyzing complex systems and supporting decision-making. Recent advances in topological signal processing highlight the importance of variables defined on topological spaces in several application domains. In such cases, the underlying topology shapes statistical relationships, limiting the expressiveness of canonical PGMs. To overcome this limitation, we introduce Colored Markov Random Fields (CMRFs), which model both conditional and marginal dependencies among Gaussian edge variables on topological spaces, with a theoretical foundation in Hodge theory. CMRFs extend classical Gaussian Markov Random Fields by including link coloring: connectivity encodes conditional independence, while color encodes marginal independence. We quantify the benefits of CMRFs through a distributed estimation case study over a physical network, comparing it with baselines with different levels of topological prior.