On the Uniqueness of Loopy Belief Propagation Fixed Points
- 1 November 2004
- journal article
- Published by MIT Press in Neural Computation
- Vol. 16 (11) , 2379-2413
- https://doi.org/10.1162/0899766041941943
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.Keywords
This publication has 7 references indexed in Scilit:
- Approximate inference in Boltzmann machinesArtificial Intelligence, 2003
- CCCP Algorithms to Minimize the Bethe and Kikuchi Free Energies: Convergent Alternatives to Belief PropagationNeural Computation, 2002
- Correctness of Belief Propagation in Gaussian Graphical Models of Arbitrary TopologyNeural Computation, 2001
- Factor graphs and the sum-product algorithmIEEE Transactions on Information Theory, 2001
- Correctness of Local Probability Propagation in Graphical Models with LoopsNeural Computation, 2000
- Turbo decoding as an instance of Pearl's "belief propagation" algorithmIEEE Journal on Selected Areas in Communications, 1998
- Local Computations with Probabilities on Graphical Structures and Their Application to Expert SystemsJournal of the Royal Statistical Society Series B: Statistical Methodology, 1988