GPU-Aware Genetic Estimation of Hidden Markov Models for Workload Classification Problems

Cuzzocrea, Alfredo; Mumolo, Enzo; Timeus, Nicola; Vercelli, Gianni

doi:10.1109/compsac.2016.123

Cited by 1 publication

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years HMM algorithms on GPUs have been implemented in various fields. A nonexhaustive list includes implementations in bioinformatics (Yao et al, 2010), speech recognition (Yu et al, 2015), a registered patent in speech matching (Chong et al, 2014) and workload classification (Cuzzocrea et al, 2016), as well as HMMer (Horn et al, 2005) an open-source project for use with protein databases. The HMM implementations are application specific often 2 with large number of states and mostly focused on increasing throughput of the Verterbi and Baum-Welch algorithms (Zhang et al, 2009;Li et al, 2009;Liu, 2009).…”

Section: Introductionmentioning

confidence: 99%

A 1000-fold Acceleration of Hidden Markov Model Fitting using Graphical Processing Units, with application to Nonvolcanic Tremor Classification

Stoltz,

Obara

et al. 2020

Preprint

View full text Add to dashboard Cite

Hidden Markov models (HMMs) are general purpose models for time-series data widely used across the sciences because of their flexibility and elegance. However fitting HMMs can often be computationally demanding and time consuming, particularly when the the number of hidden states is large or the Markov chain itself is long. Here we introduce a new Graphical Processing Unit (GPU) based algorithm designed to fit long chain HMMs, applying our approach to an HMM for nonvolcanic tremor events developed by Wang et al. (2018). Even on a modest GPU, our implementation resulted in a 1000-fold increase in speed over the standard single processor algorithm, allowing a full Bayesian inference of uncertainty related to model parameters. Similar improvements would be expected for HMM models given

show abstract

Section: Introductionmentioning

confidence: 99%