Markov Random Field Structure Estimator - Version 0.4.2 - Package mrfse

Sep 8, 2026

Upload by Hazel

Markov random field structure estimator package mrfse provides three algorithms to estimate Markovian neighborhoods from discrete samples. It includes an exact version and a simulated annealing variant of a penalized maximum conditional likelihood method, aligned with a Bayesian Information Criterion-like objective, plus a greedy approach based on referenced learning of Ising models on arbitrary graphs. The functions return estimated neighborhoods for each vertex or adjacency matrices, controlled by alphabet size, a hyperparameter tau, and a max_degree constraint. Licensed under GPL.

LanguageEnglish

CategoryTechnology

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

mrfse.ci - Bresler’s non-binary Markov random field structure estimator

Usage

Arguments

Value

References

mrfse.ci.con - Conservative approach for Bresler’s non-binary estimator

Usage

Arguments

Value

References

mrfse.ci.ncon - Non-conservative approach for Bresler’s non-binary estimator

Usage

Arguments

Value

References

Package ‘mrfse’ May 9, 2026 Title Markov Random Field Structure Estimator Version 0.4.2 Date 2024-10-21 Description Three algorithms for estimating a Markov random field structure.Two of them are an exact version and a simulated annealing version of a penalized maximum conditional likelihood method similar to the Bayesian Information Criterion. These algorithm are de scribed in Frondana (2016) .The third one is a greedy algorithm, de scribed in Bresler (2015) = 3) Maintainer Rodrigo Carvalho <[rodrigorsdc@gmail.com](rodrigorsdc@gmail.com)> LinkingTo Rcpp Imports Rcpp Depends gtools, Rfast NeedsCompilation yes Author Rodrigo Carvalho [aut, cre], Florencia Leonardi [rev, ths] Repository CRAN Date/Publication 2024-10-21 17:40:01 UTC Contents [mrfse.ci ...........................................](mrfse.ci ........................................... 2)[ 2](mrfse.ci ........................................... 2) [mrfse.ci.con .........................................](mrfse.ci.con ......................................... 3)[ 3](mrfse.ci.con ......................................... 3) [mrfse.ci.ncon ........................................](mrfse.ci.ncon ........................................ 4)[ 4](mrfse.ci.ncon ........................................ 4) [mrfse.create.sampler ....................................](mrfse.create.sampler .................................... 5)[ 5](mrfse.create.sampler .................................... 5) [mrfse.exact .........................................](mrfse.exact ......................................... 6)[ 6](mrfse.exact ......................................... 6) [mrfse.exact.con .......................................](mrfse.exact.con ....................................... 7)[ 7](mrfse.exact.con ....................................... 7) [mrfse.exact.ncon ......................................](mrfse.exact.ncon ...................................... 8)[ 8](mrfse.exact.ncon ...................................... 8) [mrfse.sa ...........................................](mrfse.sa ........................................... 9)[ 9](mrfse.sa ........................................... 9) [mrfse.sa.con .........................................](mrfse.sa.con ......................................... 10)[ 10](mrfse.sa.con ......................................... 10) [mrfse.sa.ncon ........................................](mrfse.sa.ncon ........................................ 11)[ 11](mrfse.sa.ncon ........................................ 11) [mrfse.sample ........................................](mrfse.sample ........................................ 12)[ 12](mrfse.sample ........................................ 12) [2 mrfse.ci](2 mrfse.ci) Index 13 | [mrfse.ci](mrfse.ci) | Bresler’s non-binary Markov random field structure estimator | | --- | --- | Description A greedy algorithm to estimate Markovian neighborhoods. Usage [mrfse.ci](mrfse.ci) (a_size, sample, tau, max_degree=ncol(sample)-1) Arguments a_size Size of the alphabet. sample A integer-valued matrix. Each value must belong range 0 and a_size-1. Ma trix has dimension n x V, where n is number of samples and V is number of nodes. tau A hyperparameter. See references. max_degree The maximum length of a candidate Markovian neighborhood. Must be nonnegative and less than ncol(sample) . Value A list filled with estimated Markov neighborhood for each graph vertex Author(s) Rodrigo Carvalho References Guy Bresler. 2015. Efficiently Learning Ising Models on Arbitrary Graphs. In Proceedings of the forty-seventh annual ACM symposium on Theory of Computing (STOC’15) . Association for Computing Machinery, New York, NY, USA, 771–782. DOI:[https://doi.org/10.1145/2746539.2746631](https://doi.org/10.1145/2746539.2746631) Examples library(mrfse) a_size = c(0, 1

FAQs

What problem does mrfse address?

mrfse estimates Markovian neighborhoods and Markov random field structure from integer-valued sample matrices.

Which algorithms are provided by the package?

It provides a greedy estimator plus exact and simulated annealing variants of a penalized maximum conditional likelihood method similar to a BIC-style criterion.

How does max_degree affect the results?

max_degree sets the maximum length of candidate Markovian neighborhoods and must be nonnegative and less than the number of columns in the sample matrix minus one.

Sep 8, 2026

Upload by Hazel