English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Probabilistic inference in pairwise Markov Random Fields (MRFs), i.e.computing the partition function or computing a MAP estimate of the variables,is a foundational problem in probabilistic graphical models. Semidefiniteprogramming relaxations have long been a theoretically powerful tool foranalyzing properties of probabilistic inference, but have not been practicalowing to the high computational cost of typical solvers for solving theresulting SDPs. In this paper, we propose an efficient method for computing thepartition function or MAP estimate in a pairwise MRF by instead exploiting arecently proposed coordinate-descent-based fast semidefinite solver. We alsoextend semidefinite relaxations from the typical binary MRF to the fullmulti-class setting, and develop a compact semidefinite relaxation that canagain be solved efficiently using the solver. We show that the methodsubstantially outperforms (both in terms of solution quality and speed) theexisting state of the art in approximate inference, on benchmark problems drawnfrom previous work. We also show that our approach can scale to large MRFdomains such as fully-connected pairwise CRF models used in computer vision.

Efficient semidefinite-programming-based inference for binary and multi-class MRFs