Tree consistency and bounds on the performance of the max-product algorithm and its generalizations
Statistics and Computing
Convergent Tree-Reweighted Message Passing for Energy Minimization
IEEE Transactions on Pattern Analysis and Machine Intelligence
Creating probabilistic databases from information extraction models
VLDB '06 Proceedings of the 32nd international conference on Very large data bases
Kernel-Based Learning of Hierarchical Multilabel Classification Models
The Journal of Machine Learning Research
Estimating the "Wrong" Graphical Model: Benefits in the Computation-Limited Setting
The Journal of Machine Learning Research
Walk-Sums and Belief Propagation in Gaussian Graphical Models
The Journal of Machine Learning Research
A Linear Programming Approach to Max-Sum Problem: A Review
IEEE Transactions on Pattern Analysis and Machine Intelligence
Graphical Models, Exponential Families, and Variational Inference
Foundations and Trends® in Machine Learning
An edge deletion semantics for belief propagation and its practical impact on approximation quality
AAAI'06 proceedings of the 21st national conference on Artificial intelligence - Volume 2
Revealing the structure of medical dictations with conditional random fields
EMNLP '08 Proceedings of the Conference on Empirical Methods in Natural Language Processing
Focusing generalizations of belief propagation on targeted queries
AAAI'08 Proceedings of the 23rd national conference on Artificial intelligence - Volume 2
Many-pairs mutual information for adding structure to belief propagation approximations
AAAI'08 Proceedings of the 23rd national conference on Artificial intelligence - Volume 2
Piecewise training for structured prediction
Machine Learning
Detection of Gauss-Markov random fields with nearest-neighbor dependency
IEEE Transactions on Information Theory
Convergence of min-sum message passing for quadratic optimization
IEEE Transactions on Information Theory
Notes on Cutset Conditioning on Factor Graphs with Cycles
Proceedings of the 2009 conference on Neural Nets WIRN09: Proceedings of the 19th Italian Workshop on Neural Nets, Vietri sul Mare, Salerno, Italy, May 28--30 2009
Entropy bounds for a Markov random subfield
ISIT'09 Proceedings of the 2009 IEEE international conference on Symposium on Information Theory - Volume 1
Message-passing for Graph-structured Linear Programs: Proximal Methods and Rounding Schemes
The Journal of Machine Learning Research
Reparameterization based consistent graph-structured linear programs
Proceedings of the 2010 ACM Symposium on Applied Computing
International Journal of Sensor Networks
Convergence of min-sum message-passing for convex optimization
IEEE Transactions on Information Theory
Understanding the scalability of Bayesian network inference using clique tree growth curves
Artificial Intelligence
Multifactor expectation maximization for factor graphs
ICANN'10 Proceedings of the 20th international conference on Artificial neural networks: Part III
Indoor positioning using nonparametric belief propagation based on spanning trees
EURASIP Journal on Wireless Communications and Networking - Special issue on signal processing-assisted protocols and algorithms for cooperating objects and wireless sensor networks
Max margin learning on domain-independent web information extraction
Proceedings of the 20th ACM international conference on Information and knowledge management
Boltzmann machine learning with the latent maximum entropy principle
UAI'03 Proceedings of the Nineteenth conference on Uncertainty in Artificial Intelligence
A high-speed parallel architecture for stereo matching
ISVC'06 Proceedings of the Second international conference on Advances in Visual Computing - Volume Part I
Data association based on optimization in graphical models with application to sensor networks
Mathematical and Computer Modelling: An International Journal
The Latent Maximum Entropy Principle
ACM Transactions on Knowledge Discovery from Data (TKDD)
On Learning Conditional Random Fields for Stereo
International Journal of Computer Vision
Online semantic mapping of urban environments
SC'12 Proceedings of the 2012 international conference on Spatial Cognition VIII
Journal of Computational Physics
Message-passing algorithms for quadratic minimization
The Journal of Machine Learning Research
Variational algorithms for marginal MAP
The Journal of Machine Learning Research
| Hi-index | 755.02 |
We present a tree-based reparameterization (TRP) framework that provides a new conceptual view of a large class of algorithms for computing approximate marginals in graphs with cycles. This class includes the belief propagation (BP) or sum-product algorithm as well as variations and extensions of BP. Algorithms in this class can be formulated as a sequence of reparameterization updates, each of which entails refactorizing a portion of the distribution corresponding to an acyclic subgraph (i.e., a tree, or more generally, a hypertree). The ultimate goal is to obtain an alternative but equivalent factorization using functions that represent (exact or approximate) marginal distributions on cliques of the graph. Our framework highlights an important property of the sum-product algorithm and the larger class of reparameterization algorithms: the original distribution on the graph with cycles is not changed. The perspective of tree-based updates gives rise to a simple and intuitive characterization of the fixed points in terms of tree consistency. We develop interpretations of these results in terms of information geometry. The invariance of the distribution, in conjunction with the fixed-point characterization, enables us to derive an exact expression for the difference between the true marginals on an arbitrary graph with cycles, and the approximations provided by belief propagation. More broadly, our analysis applies to any algorithm that minimizes the Bethe free energy. We also develop bounds on the approximation error, which illuminate the conditions that govern their accuracy. Finally, we show how the reparameterization perspective extends naturally to generalizations of BP (e.g., Kikuchi (1951) approximations and variants) via the notion of hypertree reparameterization.