Fast maximum a-posteriori inference on Monte Carlo state spaces

Many important algorithms for statistical inference can be expressed as a weighted maxkernel search problem. This is the case with the Viterbi algorithm for HMMs, message construction in maximum a posteriori BP (max-BP), as well as certain particlesmoothing algorithms. Previous work has focused on reducing the cost of this procedure in discrete regular grids [4]. MonteCarlo state spaces, which are vital for highdimensional inference, cannot be handled by these techniques. We present a novel dualtree based algorithm that is appliable to a wide range of kernels and shows substantial performance gains over naive computation.

[1]  William T. Freeman,et al.  Efficient Multiscale Sampling from Products of Gaussian Mixtures , 2003, NIPS.

[2]  Daniel P. Huttenlocher,et al.  Efficient Belief Propagation for Early Vision , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[3]  Hoon Kim,et al.  Monte Carlo Statistical Methods , 2000, Technometrics.

[4]  Christian P. Robert,et al.  Monte Carlo Statistical Methods , 2005, Springer Texts in Statistics.

[5]  Niclas Bergman,et al.  Recursive Bayesian Estimation : Navigation and Tracking Applications , 1999 .

[6]  Jon M. Kleinberg,et al.  Fast Algorithms for Large-State-Space HMMs with Applications to Web Usage Analysis , 2003, NIPS.

[7]  A. Doucet,et al.  Maximum a Posteriori Sequence Estimation Using Monte Carlo Particle Filters , 2001, Annals of the Institute of Statistical Mathematics.

[8]  大西 仁,et al.  Pearl, J. (1988, second printing 1991). Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan-Kaufmann. , 1994 .

[9]  James M. Coughlan,et al.  Shape Matching with Belief Propagation: Using Dynamic Quantization to Accomodate Occlusion and Clutter , 2004, 2004 Conference on Computer Vision and Pattern Recognition Workshop.

[10]  Andrew W. Moore,et al.  The Anchors Hierarchy: Using the Triangle Inequality to Survive High Dimensional Data , 2000, UAI.

[11]  Andrew W. Moore,et al.  Rapid Evaluation of Multiple Density Models , 2003, AISTATS.

[12]  Nando de Freitas,et al.  Beat Tracking the Graphical Model Way , 2004, NIPS.

[13]  Andrew W. Moore,et al.  'N-Body' Problems in Statistical Learning , 2000, NIPS.

[14]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[15]  Timothy J. Robinson,et al.  Sequential Monte Carlo Methods in Practice , 2003 .