Systems and Means of Informatics
2016, Volume 26, Issue 4, pp 60-73
APPLICATION OF THE CUDA ARCHITECTURE FOR IMPLEMENTATION OF GRID-BASED ALGORITHMS FOR THE METHOD OF MOVING SEPARATION OF MIXTURES
- A. K. Gorshenin
- V. Yu. Kuzmin
Abstract
The paper presents the implementation of the grid methods for finding maximum likelihood estimators in the mixed probability models based on the software solutions for the computations on GPUs using the NVIDIA CUDA technology. The hierarchy of programming classes is described, an approach to the initial estimation and further modification of the parametric grids is proposed, and the convergence speed and other characteristics of the developed methods are examined by the test data sets. The key characteristics of the method including the change of approximation error by the l1 metric and reducing the number of components under the iterative steps are demonstrated by the graphs.
The integration of the implemented software modules with the specialized online service for real data processing MSM Tools is also discussed.
[+] References (18)
- Chou, C.Y., Y.Y. Chuo, Y.K. Hung, and W.C. Wang. 2011. A fast forward projection using multithreads for multirays on GPUs in medical image reconstruction. Med. Phys. 38:4052-4065.
- Tahmasebi, P., M. Sahimi, G. Mariethoz, andA. Hezarkhani. 2012. A fast accelerating geostatistical simulations using graphics processing units. Comput. Geosci. 46:51-59.
- Ng, C. M. 2013. Novel hybrid GPU-CPU implementation of parallelized Monte Carlo parametric expectation maximization estimation method for population pharmacokinetic data analysis. AAPS J. 15(4): 1212-1221.
- Cardinal, P., P. Dumouchel, G. Boulianne, and M. Comeau. 2008. GPU accelerated acoustic likelihood computations. 9th Annual Conference of the International Speech Communication Association Proceedings. Brisbane, Australia. 964-967.
- Dixon, P. R., T. Oonishi, and S. Furui. 2009. Fast acoustic computations using graphics processors. IEEE Conference (International) on Acoustics, Speech and Signal Processing Proceedings. Taipei, Taiwan. 4321-4324.
- Vanek, J., J. Trmal, J. V. Psutka, and J. Psutka. 2012. Optimized acoustic likelihoods computation for NVIDIA and ATI/AMD graphics processors. IEEE Trans. Audio Speech Language Processing 20(6) :1818-1828.
- Kumar, N. S.L. P., S. Satoor, and I. Buck. 2009. Fast parallel expectation maximization for Gaussian mixture models on GPUs using CUDA. 11th IEEE Conference (International) on High Performance Computing and Communications Proceedings. Seoul, Korea. 103-109.
- Dempster, A., N. Laird, and D. Rubin. 1977. Maximum likelihood estimation from incompleted data. J. Roy. Stat. Soc. B 39(1): 1-38.
- Ayres, D.L., A. Darling, D.J. Zwickl, P. Beerli, M.T. Holder, P.O. Lewis, J. P. Huelsenbeck, F. Ronquist, D. L. Swofford, M.P. Cummings, A. Rambaut, and M.A. Suchard. 2012. BEAGLE: An application programming interface and high-performance computing library for statistical phylogenetics. Systematic Biol. 61(1): 170-173.
- Korolev, V. Yu. 2011. Veroyatnostno-statisticheskie metody dekompozitsii volatil'nosti khaoticheskikh protsessov [Probabilistic and statistical methods of decomposition of volatility of chaotic processes]. Moscow: Moscow University Publishing House. 512 p.
- Kuzmin, V.Yu., A. K Gorshenin, D. S. Ostroumov, and M. G. Uglitskaya. 2015. Application of GPU and parallel programming on grid methods. AIP Conference Proceedings 1648:250006. 4 p.
- Gorshenin, A., V. Korolev, V. Kuzmin, and A. Zeifman. 2013. Coordinate-wise versions of the grid method for the analysis of intensities of non-stationary information flows by moving separation of mixtures of gamma-distribution. 27th European Conference on Modelling and Simulation Proceedings. Dudweiler, Germany. 565-568.
- Gorshenin, A. K. 2016. Kontseptsiya onlayn-kompleksa dlya stokhasticheskogo mode- lirovaniya real'nykh protsessov [Concept of online service for stochastic modelling of real processes]. Informatika i ee Primeneniya - Inform. Appl. 10(1):72-81.
- Gorshenin, A., and V. Kuzmin. 2016. On an interface of the online system for a stochastic analysis of the varied information flows. AIP Conference Proceedings 1738: 220009. 4 p.
- Kolmogorov, A. N., and S. V. Fomin. 1976. Elementy teorii funktsiy i funktsional'nogo analiza [Elements of the theory of functions and functional analysis]. Moscow: Nauka. 544 p.
- Korolev, V. Yu., and A. L. Nazarov. 2010. Razdelenie smesey veroyatnostnykh raspre- deleniy pri pomoshchi setochnykhmetodov momentovi maksimal'nogo pravdopodobiya [Separating mixtures of probability distributions with the grid method of moments and the grid maximal likelihood method]. Avtomatika i Telemekhanika [Automation and Remote Control] 3:98-116.
- Korolev, V.Yu., A.K. Gorshenin, S. K Gulev, and K.P. Belyaev. 2015. Stati- sticheskoe modelirovanie turbulentnykh potokov tepla mezhdu okeanom i atmosferoy s pomoshch'yu metoda skol'zyashchego razdeleniya konechnykh normal'nykh smesey [Statistical modeling of air-sea turbulent heat fluxes by the method of moving separation of finite normal mixtures]. Informatika i ee Primeneniya - Inform. Appl. 9(4):3-13.
- Korolev, V. Yu., A. V. Chertok, A. Y. Korchagin, and A. I. Zeifman. 2015. Modeling high-frequency order flow imbalance by functional limit theorems for two-sided risk processes. Appl. Math. Comput. 253:224-241.
[+] About this article
Title
APPLICATION OF THE CUDA ARCHITECTURE FOR IMPLEMENTATION OF GRID-BASED ALGORITHMS FOR THE METHOD OF MOVING SEPARATION OF MIXTURES
Journal
Systems and Means of Informatics
Volume 26, Issue 4, pp 60-73
Cover Date
2016-11-30
DOI
10.14357/08696527160406
Print ISSN
0869-6527
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
CUDA; GPU; grid methods; mixed probability models; moving separation of mixtures; online software
Authors
A. K. Gorshenin  and V. Yu. Kuzmin 
Author Affiliations
Institute of Informatics Problems, Federal Research Center "Computer Science
and Control", Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Wi2Geo LLC, Moscow, Russian Federation
|