ML Anthology

On the Uniqueness of Loopy Belief Propagation Fixed Points

Tom Heskes
NeCo 2004 pp. 2379-2413
doi:10.1162/0899766041941943 /neco/2004/heskes2004neco-uniqueness/

Abstract

We derive sufficient conditions for the uniqueness of loopy belief propagation fixed points. These conditions depend on both the structure of the graph and the strength of the potentials and naturally extend those for convexity of the Bethe free energy. We compare them with (a strengthened version of) conditions derived elsewhere for pairwise potentials. We discuss possible implications for convergent algorithms, as well as for other approximate free energies.

PDF NeCo Semantic Scholar

Cite

Text

Heskes. "On the Uniqueness of Loopy Belief Propagation Fixed Points." Neural Computation, 2004. doi:10.1162/0899766041941943

Markdown

[Heskes. "On the Uniqueness of Loopy Belief Propagation Fixed Points." Neural Computation, 2004.](https://mlanthology.org/neco/2004/heskes2004neco-uniqueness/) doi:10.1162/0899766041941943

BibTeX

@article{heskes2004neco-uniqueness,
  title     = {{On the Uniqueness of Loopy Belief Propagation Fixed Points}},
  author    = {Heskes, Tom},
  journal   = {Neural Computation},
  year      = {2004},
  pages     = {2379-2413},
  doi       = {10.1162/0899766041941943},
  volume    = {16},
  url       = {https://mlanthology.org/neco/2004/heskes2004neco-uniqueness/}
}