Validity Estimates for Loopy Belief Propagation on Binary Real-world Networks
Mooij, Joris M., Kappen, Hilbert J.
–Neural Information Processing Systems
We introduce a computationally efficient method to estimate the validity of the BP method as a function of graph topology, the connectivity strength, frustration and network size. We present numerical results that demonstrate the correctness of our estimates for the uniform random model and for a real-world network ("C.
Neural Information Processing Systems
Dec-31-2005