SoC Processor Discovery for Program Execution Matching Using Unsupervised Machine Learning

Avi Bleiweiss

2014

Abstract

The fast cadence for evolving mobile compute systems, often extends their default processor configuration by incorporating task specific, companion cores. In this setting, the problem of matching a compute program to efficiently execute on a dynamically selected processor, poses a considerable challenge to employing traditional compiler technology. Rather, we propose an unsupervised machine learning methodology that mines a large data corpus of unlabeled compute programs, with the objective to discover optimal program-processor relations. In our work, we regard a compute program as a text document, comprised of a linear sequence of bytecode mnemonics, and further transformed into an effective representation of a bag of instruction term frequencies. Respectively, a set of concise instruction vectors is forwarded onto a finite mixture model, to identify unsolicited cluster patterns of source-target compute pairings, using the expectation-maximization algorithm. For classification, we explore k-nearest neighbor and ranked information retrieval methods, and evaluate our system by simultaneously varying the dimensionality of the training set and the SoC processor formation. We report robust performance results on both the discovery of relational clusters and feature matching.

Paper Citation

in Harvard Style

Bleiweiss A. (2014). SoC Processor Discovery for Program Execution Matching Using Unsupervised Machine Learning . In Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2014) ISBN 978-989-758-048-2, pages 192-201. DOI: 10.5220/0005070301920201

in Bibtex Style

@conference{kdir14,
author={Avi Bleiweiss},
title={SoC Processor Discovery for Program Execution Matching Using Unsupervised Machine Learning},
booktitle={Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2014)},
year={2014},
pages={192-201},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005070301920201},
isbn={978-989-758-048-2},
}

in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2014)
TI - SoC Processor Discovery for Program Execution Matching Using Unsupervised Machine Learning
SN - 978-989-758-048-2
AU - Bleiweiss A.
PY - 2014
SP - 192
EP - 201
DO - 10.5220/0005070301920201