Skip to Main content Skip to Navigation
Conference papers

Parameter Learning for Log-supermodular Distributions

Tatiana Shpakova 1, 2 Francis Bach 1, 2
1 SIERRA - Statistical Machine Learning and Parsimony
DI-ENS - Département d'informatique de l'École normale supérieure, CNRS - Centre National de la Recherche Scientifique, Inria de Paris
Abstract : We consider log-supermodular models on binary variables, which are probabilistic models with negative log-densities which are submodular. These models provide probabilistic interpretations of common combinatorial optimization tasks such as image segmentation. In this paper, we focus primarily on parameter estimation in the models from known upper-bounds on the intractable log-partition function. We show that the bound based on separable optimization on the base polytope of the submodular function is always inferior to a bound based on " perturb-and-MAP " ideas. Then, to learn parameters, given that our approximation of the log-partition function is an expectation (over our own randomization), we use a stochastic subgradient technique to maximize a lower-bound on the log-likelihood. This can also be extended to conditional maximum likelihood. We illustrate our new results in a set of experiments in binary image denoising, where we highlight the flexibility of a probabilistic model to learn with missing data.
Complete list of metadata

Cited literature [24 references]  Display  Hide  Download

https://hal.inria.fr/hal-01354789
Contributor : Tatiana Shpakova <>
Submitted on : Thursday, November 3, 2016 - 6:34:34 PM
Last modification on : Tuesday, May 4, 2021 - 2:06:02 PM
Long-term archiving on: : Saturday, February 4, 2017 - 2:25:18 PM

File

nips_2016.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01354789, version 2
  • ARXIV : 1608.05258

Collections

Citation

Tatiana Shpakova, Francis Bach. Parameter Learning for Log-supermodular Distributions. NIPS 2016 - Thirtieth Annual Conference on Neural Information Processing System, Dec 2016, Barcelona, Spain. ⟨hal-01354789v2⟩

Share

Metrics

Record views

305

Files downloads

803