Work

AD3 : Alternating Directions Dual Decomposition for MAP Inference in Graphical Models

Year: 2018

Type: article

Abstract: We present AD3, a new algorithm for approximate maximum a posteriori (MAP) inference on factor graphs, based on the alternating directions method of multipliers. Like other dual decomposition algorithms, AD3 has a modular architecture, where local subproblems are solved independently, and their solutions are gathered to compute a global update. The key characteristic of AD3 is that each local subproblem has a quadratic regularizer, leading to faster convergence, both theoretically and in practice. We provide closed-form solutions for these AD3 subproblems for binary pairwise factors and factors imposing first-order logic constraints. For arbitrary factors (large or combinatorial), we introduce an active set method which requires only an oracle for computing a local MAP configuration, making AD3 applicable to a wide range of problems. Experiments on synthetic and real-world problems show that AD3 compares favorably with the state-of-the-art.

DOI: 10.1184/r1/6473018

Authors: Martins, Andre F.T., Figueiredo, Mario A. T., Aguiar, Pedro M.Q., A. Smith, Noah, P Xing, Eric

Institutions: Instituto de Telecomunicações, University of Lisbon, Instituto Superior Técnico, Carnegie Mellon University

Language: English

Topic: Bayesian Modeling and Causal Inference

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W237364718

FWCI: 3.267

Cites: 75

Cited by: 47

Related to: 10