ICML 2019oral47 citations

Optimal Continuous DR-Submodular Maximization and Applications to Provable Mean Field Inference

Yatao Bian, Joachim Buhmann, Andreas Krause

Abstract

Mean field inference for discrete graphical models is generally a highly nonconvex problem, which also holds for the class of probabilistic log-submodular models. Existing optimization methods, e.g., coordinate ascent algorithms, typically only find local optima. In this work we propose provable mean filed methods for probabilistic log-submodular models and its posterior agreement (PA) with strong approximation guarantees. The main algorithmic technique is a new Double Greedy scheme, termed DR-DoubleGreedy, for continuous DR-submodular maximization with box-constraints. It is a one-pass algorithm with linear time complexity, reaching the optimal 1/2 approximation ratio, which may be of independent interest. We validate the superior performance of our algorithms against baselines on both synthetic and real-world datasets.

BibTeX
@InProceedings{pmlr-v97-bian19a,
  title = 	 {Optimal Continuous {DR}-Submodular Maximization and Applications to Provable Mean Field Inference},
  author =       {Bian, Yatao and Buhmann, Joachim and Krause, Andreas},
  booktitle = 	 {Proceedings of the 36th International Conference on Machine Learning},
  pages = 	 {644--653},
  year = 	 {2019},
  editor = 	 {Chaudhuri, Kamalika and Salakhutdinov, Ruslan},
  volume = 	 {97},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {09--15 Jun},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v97/bian19a/bian19a.pdf},
  url = 	 {https://proceedings.mlr.press/v97/bian19a.html},
  abstract = 	 {Mean field inference for discrete graphical models is generally a highly nonconvex problem, which also holds for the class of probabilistic log-submodular models. Existing optimization methods, e.g., coordinate ascent algorithms, typically only find local optima. In this work we propose provable mean filed methods for probabilistic log-submodular models and its posterior agreement (PA) with strong approximation guarantees. The main algorithmic technique is a new Double Greedy scheme, termed DR-DoubleGreedy, for continuous DR-submodular maximization with box-constraints. It is a one-pass algorithm with linear time complexity, reaching the optimal 1/2 approximation ratio, which may be of independent interest. We validate the superior performance of our algorithms against baselines on both synthetic and real-world datasets.}
}