论文信息 - Anytime marginal maximum a posteriori inference

Anytime marginal maximum a posteriori inference

This paper presents a new anytime algorithm for the marginal MAP problem in graphical models of bounded treewidth. We show asymptotic convergence and theoretical error bounds for any fixed step. Experiments show that it compares well to a state-of-the-art systematic search algorithm.

Denis Deratani Mauá | Cassio Polpo de Campos

[1] Adnan Darwiche,et al. Solving MAP Exactly using Systematic Search , 2002, UAI.

[2] Nir Friedman,et al. Probabilistic Graphical Models - Principles and Techniques , 2009 .

[3] Cassio Polpo de Campos. Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence New Complexity Results for MAP in Bayesian Networks , 2022 .

[4] Qiang Liu,et al. Variational algorithms for marginal MAP , 2011, J. Mach. Learn. Res..

[5] A. Darwiche,et al. Complexity Results and Approximation Strategies for MAP Explanations , 2011, J. Artif. Intell. Res..

[6] Christopher A. Meek,et al. Approximating Max‐Sum‐Product Problems using Multiplicative Error Bounds , 2011 .

[7] Hal Daumé,et al. Message-Passing for Approximate MAP Inference with Latent Variables , 2011, NIPS.

[8] Changhe Yuan,et al. Annealed MAP , 2004, UAI.