Dual Decomposition for Joint Discrete-Continuous Optimization

@inproceedings{Zach2013DualDF,
  title={Dual Decomposition for Joint Discrete-Continuous Optimization},
  author={Christopher Zach},
  booktitle={AISTATS},
  year={2013}
}
We analyse convex formulations for combined discrete-continuous MAP inference using the dual decomposition method. As a consquence we can provide a more intuitive derivation for the resulting convex relaxation than presented in the literature. Further, we show how to strengthen the relaxation by reparametrizing the potentials, hence convex relaxations for discrete-continuous inference does not share an important feature of LP relaxations for discrete labeling problems: incorporating unary… CONTINUE READING
7 Citations
25 References
Similar Papers

References

Publications referenced by this paper.
Showing 1-10 of 25 references

Lay - ered image motion with explicit occlusions , temporal consistency , and depth ordering Repre - senting moving images with layers

  • T. Meltzer, A. Globerson, T. Jaakkola, Y. Weiss, E. Sudderth, M. Black
  • IEEE Trans . Image Proc .
  • 2011

Optimization for Machine Learning, chapter Introduction to Dual Decomposition for Inference

  • D. Sontag, A. Globerson, T. Jaakkola
  • 2011

Similar Papers

Loading similar papers…