ABSTRACT
In our work with scientists, we find that Gaussian Mixture Modeling is a common type of analysis applied to increasingly large datasets. We implement this algorithm in the Myria shared-nothing relational data management system, which performs the computation in memory. We study resulting memory utilization challenges and implement several optimizations that yield an efficient and scalable solution. Empirical evaluations on large astronomy and oceanography datasets confirm that our Myria approach scales well and performs up to an order of magnitude faster than Hadoop.
- R. Adams. Computing log-sum-exp, Jan. 2013.Google Scholar
- A. Ailamaki, D. J. DeWitt, M. D. Hill, and M. Skounakis. Weaving relations for cache performance. In Proc. of VLDB, pages 169--180, 2001. Google ScholarDigital Library
- D. Arthur and S. Vassilvitskii. k-means++: the advantages of careful seeding. In Proc. of SODA, pages 1027--1035, 2007. Google ScholarDigital Library
- T. Benaglia, D. Chauveau, D. R. Hunter, and D. Young. mixtools: An R package for analyzing finite mixture models. Journal of Statistical Software, 32(6):1--29, 2009.Google ScholarCross Ref
- L. S. Blackford, J. Choi, A. Cleary, E. D'Azevedo, J. Demmel, I. Dhillon, J. Dongarra, S. Hammarling, G. Henry, A. Petitet, K. Stanley, D. Walker, and R. C. Whaley. ScaLAPACK Users' Guide. Society for Industrial and Applied Mathematics, Philadelphia, PA, 1997. Google ScholarDigital Library
- P. G. Brown. Overview of scidb: large scale array storage, processing and analysis. In Proc. of SIGMOD, pages 963--968, 2010. Google ScholarDigital Library
- C. Chu et al. Map-reduce for machine learning on multicore. In Proc. of NIPS, pages 281--288, 2006.Google Scholar
- HP Distributed R. http://www.distributedr.org.Google Scholar
- C. Fraley and A. E. Raftery. Model-based clustering, discriminant analysis and density estimation. Journal of the American Statistical Association, 97:611--631, 2002.Google ScholarCross Ref
- Apache Hadoop. http://hadoop.apache.org.Google Scholar
- D. Halperin et al. Demonstration of the Myria big data management service. In Proc. of SIGMOD, pages 881--884, 2014. Google ScholarDigital Library
- J. Hicklin, C. Moler, P. Webb, R. F. Boisvert, B. Miller, R. Pozo, and K. Remington. Jama: A Java matrix package. URL: http://math. nist.gov/javanumerics/jama, 2000.Google Scholar
- J. Hyrkas, D. Halperin, and B. Howe. Time-varying clusters in large-scale flow cytometry. In Proc. of AAAI, pages 4022--4023, 2015.Google Scholar
- jblas: Linear Algebra for Java. http://jblas.org.Google Scholar
- LAPACK: âĂL' Linear Algebra PACKage. http://www.netlib.org/lapack.Google Scholar
- Y. Low et al. Distributed GraphLab: A framework for machine learning in the cloud. PVLDB, 5(8):716--727, 2012. Google ScholarDigital Library
- Apache Mahout. http://mahout.apache.org.Google Scholar
- Spark Machine Learning Library (MLlib). http://spark.apache.org/mllib.Google Scholar
- K. P. Murphy. Machine Learning: A Probabilistic Perspective. The MIT Press, 2012. Google ScholarDigital Library
- Myria: Big Data as a Service. http://myria.cs.washington.edu.Google Scholar
- F. Pedregosa et al. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12:2825--2830, 2011. Google ScholarDigital Library
- Sloan Digital Sky Survey III: DR 10. http://www.sdss3.org/dr10/.Google Scholar
- J. Swalwell, F. Ribalet, and E. Armbrust. Seaflow: A novel underway flow-cytometer for continuous observations of phytoplankton in the ocean. Limnology & Oceanography Methods, 9:466--477, 2011.Google ScholarCross Ref
- Wide-field Infrared Survey Explorer. http://www.nasa.gov/mission_pages/WISE/main/index.html.Google Scholar
- M. Zaharia et al. Resilient distributed datasets: A fault-tolerant abstraction for in-memory cluster computing. In Proc. of USENIX, pages 15--28, 2012. Google ScholarDigital Library
Recommendations
Gaussian mixture density modeling, decomposition, and applications
We present a new approach to the modeling and decomposition of Gaussian mixtures by using robust statistical methods. The mixture distribution is viewed as a contaminated Gaussian density. Using this model and the model-fitting (MF) estimator, we propose ...
PARAFAC-Based Blind Identification of Underdetermined Mixtures Using Gaussian Mixture Model
This paper presents a novel algorithm, named GMM-PARAFAC, for blind identification of underdetermined instantaneous linear mixtures. The GMM-PARAFAC algorithm uses Gaussian mixture model (GMM) to model non-Gaussianity of the independent sources. We show ...
A BYY Split-and-Merge EM Algorithm for Gaussian Mixture Learning
ISNN '08: Proceedings of the 5th international symposium on Neural Networks: Advances in Neural NetworksGaussian mixture is a powerful statistic tool and has been widely used in the fields of information processing and data analysis. However, its model selection, i.e., the selection of number of Gaussians in the mixture, is still a difficult problem. ...
Comments