Approximate Inference by Kullback-Leibler Tensor Belief Propagation

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Downloads (Pure)

Abstract

Probabilistic programming provides a structured approach to signal processing algorithm design. The design task is formulated as a generative model, and the algorithm is derived through automatic inference. Efficient inference is a major challenge; e.g., the Shafer-Shenoy algorithm (SS) performs badly on models with large treewidth, which arise from various real-world problems. We focus on reducing the size of discrete models with large treewidth by storing intermediate factors in compressed form, thereby decoupling the variables through conditioning on introduced weights. This work proposes pruning of these weights using Kullback-Leibler divergence. We adapt a strategy from the Gaussian mixture reduction literature, leading to Kullback-Leibler Tensor Belief Propagation (KL-TBP), in which we use agglomerative hierarchical clustering to subsequently merge pairs of weights. Experiments using benchmark problems show KL-TBP consistently achieves lower approximation error than existing methods with competitive runtime.
Original languageEnglish
Title of host publication2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Place of PublicationBarcelona, Spain
PublisherInstitute of Electrical and Electronics Engineers
Pages5850-5854
Number of pages5
ISBN (Electronic)978-1-5090-6631-5
ISBN (Print)978-1-5090-6632-2
DOIs
Publication statusPublished - 14 May 2020
Event2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2020) - Virtual, Barcelona, Spain
Duration: 4 May 20208 May 2020
https://2020.ieeeicassp.org/

Conference

Conference2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2020)
Abbreviated titleICASSP 2020
CountrySpain
CityBarcelona
Period4/05/208/05/20
Internet address

Keywords

  • Approximation algorithms
  • Bayes methods
  • Dimensionality reduction
  • Tensors

Fingerprint

Dive into the research topics of 'Approximate Inference by Kullback-Leibler Tensor Belief Propagation'. Together they form a unique fingerprint.

Cite this