EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Compressed computations using wavelets for hidden Markov models with continuous observations

Luca Bello, John Wiedenhöft iD, Alexander Schliep iD

PLOS ONE · 2023

DOI: 10.1371/journal.pone.0286074 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, 2023. Verfasst von Luca Bello und 2 weiteren Autor:innen. Beteiligte Einrichtung: Computer Science and Engineering, University of Gothenburg, Chalmers, Gothenburg, Sweden. Hauptthemen laut MEDLINE: Algorithms. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Compression as an accelerant of computation is increasingly recognized as an important component in engineering fast real-world machine learning methods for big data; c.f., its impact on genome-scale approximate string matching. Previous work showed that compression can accelerate algorithms for Hidden Markov Models (HMM) with discrete observations, both for the classical frequentist HMM algorithms—Forward Filtering, Backward Smoothing and Viterbi—and Gibbs sampling for Bayesian HMM. For Bayesian HMM with continuous-valued observations, compression was shown to greatly accelerate computations for specific types of data. For instance, data from large-scale experiments interrogating structural genetic variation can be assumed to be piece-wise constant with noise, or, equivalently, data generated by HMM with dominant self-transition probabilities. Here we extend the compressive computation approach to the classical frequentist HMM algorithms on continuous-valued observations, providing the first compressive approach for this problem. In a large-scale simulation study, we demonstrate empirically that in many settings compressed HMM algorithms very clearly outperform the classical algorithms with no, or only an insignificant effect, on the computed probabilities and infered state paths of maximal likelihood. This provides an efficient approach to big data computations with HMM. An open-source implementation of the method is available from https://github.com/lucabello/wavelet-hmms .

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

AlgorithmsMarkov ChainsBayes TheoremProbabilityComputer Simulation

Autor:innen und Institutionen

  • Computer Science and Engineering, University of Gothenburg, Chalmers, Gothenburg, Sweden.
  • Scientific Core Facility Medical Biometry and Statistical Bioinformatics, University Medical Center Göttingen, Göttingen, Germany.
  • Faculty of Health Sciences, B-TU Cottbus-Senftenberg, Cottbus, Germany.

ORCID-Profile: John Wiedenhöft, Alexander Schliep

Zitieren

APA

Bello, L., Wiedenhöft, J., & Schliep, A. (2023). Compressed computations using wavelets for hidden Markov models with continuous observations. PLOS ONE. https://doi.org/10.1371/journal.pone.0286074

Vancouver

Bello L, Wiedenhöft J, Schliep A. Compressed computations using wavelets for hidden Markov models with continuous observations. PLOS ONE. 2023. doi:10.1371/journal.pone.0286074

BibTeX
@article{Bello2023,
  author = {Bello, Luca and Wiedenhöft, John and Schliep, Alexander},
  title = {Compressed computations using wavelets for hidden Markov models with continuous observations},
  journal = {PLOS ONE},
  year = {2023},
  doi = {10.1371/journal.pone.0286074}
}
Kostenlos starten →