Informatics and Applications
2021, Volume 15, Issue 3, pp 63-74METHOD FOR IMPROVING ACCURACY OF NEURAL NETWORK FORECASTS BASED ON PROBABILITY MIXTURE MODELS AND ITS IMPLEMENTATION AS A DIGITAL SERVICE
- A. K. Gorshenin
- V. Yu. Kuzmin
Abstract
A method aimed at improving the forecasting accuracy is presented. It uses a combination ofclassical probabilistic-statistical models and neural networks. Moments of mathematical models are used as a nontrivial expansion ofthe feature space. The efficiency ofthe proposed approach is demonstrated by the analysis ofseveral experimental data ensembles of the L-2M stellarator. Error decrease is especially noticeable when using the moments of the statistical models based on the increments of the initial observed data. To implement the methods of statistical analysis and the proposed machine learning algorithms, a digital service has been created. Its architecture and capabilities are also outlined.[+] References (28)
- 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.
- Gorshenin, A. K., V. Yu. Korolev, and A. I. Zeifman. 2020. Modeling particle size distribution in lunar regolith via a central limit theorem for random sums. Mathematics 8(9): 1409. 24 p.
- Korolev, V. Yu., and A. K. Gorshenin. 2020. Probability models and statistical tests for extreme precipitation based on generalized negative binomial distributions. Mathematics 8(4):604.
- Chandrashekar, G., andF Sahin. 2014. Asurvey on feature selection methods. Comput. Electr. Eng. 40(1):16-28.
- Gorshenin, A., and V. Kuzmin. 2021. On feature expansion with finite normal mixture models in machine learning. Advances in intelligent systems, computer science and digital economics II. Eds. Zh. Hu, S. Petoukhov, and M. He. Advances in intelligent systems and computing ser. Springer. 1402:82-90.
- Batanov, G. M., V. D. Borzosekov, A. K. Gorshenin, N. K. Kharchev, V. Yu. Korolev, and K. A. Sarskyan. 2019. Evolution of statistical properties of microturbulence during transient process under electron cyclotron resonance heating of the L-2M stellarator plasma. Plasma Phys. Contr. F. 61(7):075006. 7 p.
- Meneghini, O., C.J. Luna, S. P. Smith, and L. L. Lao. 2014. Modeling oftransport phenomena in tokamak plasmas with neural networks. Phys. Plasmas 21(6):060702. 4 p.
- Raja, M.A. Z., F. H. Shah, M. Tariq, I. Ahmad, and S.U. Ahmad. 2018. Design of artificial neural network models optimized with sequential quadratic programming to study the dynamics of nonlinear Troesch's problem arising in plasma physics. Neural Computing Applications 29(6):83-109.
- Mesbah, A., and D. B. Graves. 2019. Machine learning for modeling, diagnostics, and control of non-equilibrium plasmas. J. Phys. DAppl. Phys. 52(30):30LT02. 9 p.
- Narita, E., M. Honda, M. Nakata, M. Yoshida, N. Hayashi, and H. Takenaga. 2019. Neural-network- based semi-empirical turbulent particle transport modelling founded on gyrokinetic analyses of JT-60U plasmas. Nucl. Fusion 59(10):106018. 17 p.
- Parsons, M. S. 2017. Interpretation of machine-learning- based disruption models for plasma control. Plasma Phys. Contr. F. 59(8):085001. 5 p.
- Kates-Harbeck, J., A. Svyatkovskiy, and W Tang. 2019. Predicting disruptive instabilities in controlled fusion plasmas through deep learning. Nature 568(7753):526-531.
- Batanov, G. M., M.S. Berezhetskii, V. D. Borzosekov, et al. 2017. Reaction of turbulence at the edge and in the center of the plasma column to pulsed impurity injection caused by the sputtering of the wall coating in L-2M stellarator. Plasma Phys. Rep. 43(8):818-823.
- Gorshenin, A. K. 2016. Kontseptsiya onlayn-kompleksa dlya stokhasticheskogo modelirovaniya real'nykh protsessov [Concept of online service for stochastic modeling of real processes]. Informatika i ee Primeneniya - Inform. Appl. 10(1):72-81. doi: 10.14357/19922264160107.
- Buduma, N. 2017. Fundamentals of deep learning: De-signing next-generation machine intelligence algorithms. Sebastopol, CA: O'Reilly Media. 298 p.
- Srivastava, N., G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov. 2014. Dropout: A simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15:1929-1958.
- Gorshenin, A. K., and V. Yu. Kuzmin. 2019. Improved architecture of feedforward neural networks to increase accuracy of predictions for moments of finite normal mixtures. Pattern Recognition Image Analysis 29(1):68-77.
- Gorshenin, A. K. 2015. On implementation of EM-type algorithms in the stochastic models for a matrix computing on GPU. AIPConf. Proc. 1648:250008. 4 p.
- Liu, C., H.-C. Li, K. Fu, F. Zhang, M. Datcu, and W J. Emery. 2019. A robust EM clustering algorithm for Gaussian mixture models. Pattern Recogn. 87:269-284.
- Wu, D., and J. Ma. 2019. An effective EM algorithm for mixtures of Gaussian processes via the MCMC sampling and approximation. Neurocomputing 331:366-374.
- Zeller, C.B., C.R.B. Cabral, V. H. Lachos, and L. Benites. 2019. Finite mixture of regression models for censored data based on scale mixtures of normal distributions. Adv. Data Anal. Classi. 13(1):89-116.
- Gorshenin, A., and V. Kuzmin. 2015. Online system for the construction of structural models of information flows. 7th Congress (International) on Ultra Modern Telecommunications and Control Systems and Workshops Proceedings. Piscataway, NJ: IEEE. 216-219.
- Gorshenin, A., and V. Kuzmin. 2016. On an interface of the online system for a stochastic analysis of the varied information flows. AIPConf. Proc. 1738:220009. 4 p.
- Gorshenin, A. K. 2017. O nekotorykh matematicheskikh i programmnykh metodakh postroeniya strukturnykh modeley informatsionnykh potokov [On some mathematical and programming methods for construction of structural models of information flows]. Informatika i ee Primeneniya - Inform. Appl. 11(1):58-68. doi: 10.14357/ 19922264170105.
- Gorshenin, A. K., and V. Yu. Kuzmin. 2017. Research support system for stochastic data processing. Pattern Recognition Image Analysis 27(3):518-524.
- Bergstra, J., and Y. Bengio. 2012. Random search for hyper-parameter optimization. J. Mach. Learn. Res. 13:281-305.
- Aymar, R., P Barabaschi, and Y. Shimomura. 2002. The ITER design. Plasma Phys. Contr. F. 44(5):519-565.
- Greff, K., R. K. Srivastava, J. Koutnik, B. R. Steunebrink, andJ. Schmidhuber. 2017. LSTM: A search space odyssey. IEEE T. Neur. Net. Lear. 28(10):2222-2232.
[+] About this article
Title
METHOD FOR IMPROVING ACCURACY OF NEURAL NETWORK FORECASTS BASED ON PROBABILITY MIXTURE MODELS AND ITS IMPLEMENTATION AS A DIGITAL SERVICEJournal
Informatics and Applications2021, Volume 15, Issue 3, pp 63-74
Cover Date
2021-09-30DOI
10.14357/19922264210309Print ISSN
1992-2264Publisher
Institute of Informatics Problems, Russian Academy of SciencesAdditional Links
Key words
neural networks; finite normal mixtures; probability models; forecasting; digital service; high- performance computing; turbulence plasma; stellaratorAuthors
A. K. Gorshenin
and V. Yu. Kuzmin
Author Affiliations
Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Faculty of Space Research, M.V. Lomonosov Moscow State University, 1-52 Leninskie Gory, GSP-1, Moscow 119991, Russian Federation