Creating and sharing knowledge for telecommunications

Augmented Dual Decomposition for MAP Inference

Martins, A. ; Figueiredo, M. A. T. ; Aguiar, P. ; Smith, N. A. ; Xing, E. P.

Augmented Dual Decomposition for MAP Inference, Proc NIPS Workshop in Optimization for Machine Learning, Whistler, Canada, Vol. ?, pp. ? - ?, December, 2010.

Digital Object Identifier:


In this paper, we propose combining augmented Lagrangian optimization

with the dual

decomposition method to obtain a fast algorithm for approximate MAP

(\emph{maximum a posteriori}) inference on factor graphs. We also show how the

proposed algorithm can efficiently handle problems with (possibly global)

structural constraints. The experimental results reported testify for the

state-of-the-art performance of the proposed approach.