Generalized Majorization-Minimization

@article{Parizi2015GeneralizedM,
  title={Generalized Majorization-Minimization},
  author={Sobhan Naderi Parizi and Kun He and Stan Sclaroff and Pedro F. Felzenszwalb},
  journal={CoRR},
  year={2015},
  volume={abs/1506.07613}
}
Non-convex optimization is ubiquitous in machine learning. The MajorizationMinimization (MM) procedure systematically optimizes non-convex functions through an iterative construction and optimization of upper bounds on the objective function. The bound at each iteration is required to touch the objective function at the optimizer of the previous bound. We show that this touching constraint is unnecessary and overly restrictive. We generalize MM by relaxing this constraint, and propose a new… CONTINUE READING
Recent Discussions
This paper has been referenced on Twitter 21 times over the past 90 days. VIEW TWEETS

Citations

Publications citing this paper.

References

Publications referenced by this paper.
Showing 1-10 of 26 references

, and Zoubin Ghahramani . On the convergence of bound optimization algorithms

  • Sam Roweis Ruslan Salakhutdinov
  • Towards automatic bounding box annotations from…
  • 2015

Similar Papers

Loading similar papers…