A kernel support vector machine trained using approximate global and exhaustive local sampling

Benjamin Bryant, Hamed Sari-Sarraf, Rodney Long, Sameer Antani

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

AGEL-SVM is an extension to a kernel Support Vector Machine (SVM) and is designed for distributed computing using Approximate Global Exhaustive Local sampling (AGEL)-SVM. The dual form of SVM is typically solved using sequential minimal optimization (SMO) which iterates very fast if the full kernel matrix can fit in a computer’s memory. AGEL-SVM aims to partition the feature space into sub problems such that the kernel matrix per problem can fit in memory by approximating the data outside each partition. AGEL-SVM has similar Cohen’s Kappa and accuracy metrics as the underlying SMO implementation. AGEL-SVM’s training times greatly decreased when running on a 128 worker MATLAB pool on Amazon’s EC2. Predictor evaluation times are also faster due to a reduction in support vectors per partition.

Original languageEnglish
Title of host publicationBDCAT 2017 - Proceedings of the 4th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies
PublisherAssociation for Computing Machinery, Inc
Pages267-268
Number of pages2
ISBN (Electronic)9781450355490
DOIs
StatePublished - Dec 5 2017
Event4th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies, BDCAT 2017 - Austin, United States
Duration: Dec 5 2017Dec 8 2017

Publication series

NameBDCAT 2017 - Proceedings of the 4th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies

Conference

Conference4th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies, BDCAT 2017
CountryUnited States
CityAustin
Period12/5/1712/8/17

Keywords

  • AGEL
  • AGEL-SVM
  • AMAZON
  • Distributed
  • EC2
  • Kernel
  • MATLAB
  • SVM

Fingerprint Dive into the research topics of 'A kernel support vector machine trained using approximate global and exhaustive local sampling'. Together they form a unique fingerprint.

  • Cite this

    Bryant, B., Sari-Sarraf, H., Long, R., & Antani, S. (2017). A kernel support vector machine trained using approximate global and exhaustive local sampling. In BDCAT 2017 - Proceedings of the 4th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (pp. 267-268). (BDCAT 2017 - Proceedings of the 4th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies). Association for Computing Machinery, Inc. https://doi.org/10.1145/3148055.3149206