Back to Search
Start Over
Approximate Inference by Kullback-Leibler Tensor Belief Propagation
- Source :
- 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 5850-5854, STARTPAGE=5850;ENDPAGE=5854;TITLE=2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), ICASSP
- Publication Year :
- 2020
- Publisher :
- Institute of Electrical and Electronics Engineers, 2020.
-
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.
- Subjects :
- Kullback–Leibler divergence
Computer science
Dimensionality reduction
Gaussian
05 social sciences
Probabilistic logic
Approximation algorithm
Inference
Tensors
Belief propagation
050105 experimental psychology
Approximation algorithms
Bayes methods
Treewidth
03 medical and health sciences
symbols.namesake
Approximate inference
Generative model
0302 clinical medicine
symbols
0501 psychology and cognitive sciences
Algorithm
030217 neurology & neurosurgery
Subjects
Details
- Language :
- English
- Database :
- OpenAIRE
- Journal :
- 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 5850-5854, STARTPAGE=5850;ENDPAGE=5854;TITLE=2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), ICASSP
- Accession number :
- edsair.doi.dedup.....01eaef9f2117be775d445750c2b31fc6