Scaling the iHMM: Parallelization versus Hadoop

This paper compares parallel and distributed implementations of an iterative, Gibbs sampling, machine learning algorithm. Distributed implementations run under Hadoop on facility computing clouds. The probabilistic model under study is the infinite HMM, in which parameters are learnt using an instan...

Full description

Saved in:
Bibliographic Details
Main Authors: Bratières, Sébastien, van Gael, Jurgen, Vlachos, Andreas, Ghahramani, Zoubin
Format: Conference Proceeding
Language:English
Subjects:
Computational modeling
Data models
Hidden Markov models
Machine learning
Markov processes
Probabilistic logic
Tagging
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper compares parallel and distributed implementations of an iterative, Gibbs sampling, machine learning algorithm. Distributed implementations run under Hadoop on facility computing clouds. The probabilistic model under study is the infinite HMM, in which parameters are learnt using an instance blocked Gibbs sampling, with a step consisting of a dynamic program. We apply this model to learn part-of-speech tags from newswire text in an unsupervised fashion. However our focus here is on runtime performance, as opposed to NLP-relevant scores, embodied by iteration duration, ease of development, deployment and debugging.
DOI:10.1109/CIT.2010.223