Научная электронная библиотека ULRICH'S Periodicals Directory

Сибирский государственный университет
науки и технологий имени академика М.Ф. Решетнева

Сибирский аэрокосмический журнал
ISSN 2712-8970 (Print) ISSN 2782-5760 (on-line)

Сведения об образовательной организации

UDK 519.7; 519.66; 57.087.1, 612.087.1 Vestnik SibGAU. 2014, No. 3(55), P. 68–72
ASSESSMENT OF STABILITY LEARNING ALGORITHMS LARGE ARTIFICIAL NEURAL NETWORKS OF BIOMETRIC APPLICATION
S. V. Kachalin
Penza State University 40, Krasnaya str., Penza, 440026, Russian Federation Е-mail: cnit@pnzgu.ru
The problem of increasing the speed, stability and efficiency of known and emerging algorithms for training large artificial neural networks is considered. To solve this problem it is proposed to create the most balanced multi-criteria performance comparison of different learning algorithms, enter a numerical estimate of the learning parameters and systematically carry out work to improve them. It is argued that resistance training of neural networks is an analogue of the assessment procedure of the condition number of matrices of linear algebra. We propose a measure of stability training, given the connection between this indicator with the computational complexity of learning algorithms, as well as with a number of examples on which the algorithm is efficient. We discuss the effect of increasing the stability by supplementing the training sample synthetic image examples of "your". The optimum amount of additional synthetic examples, the addition of which can reduce the number of examples in the training set is provided. A block diagram of a promising means of training neural network converters, which introduced the third machine reproduction examples image of "Its" is given. To control the number of synthetic examples, it is proposed to create a table of restrictions for the first machines based on expert opinion.
biometrics, neural network, learning, pattern recognition, statistics, probability, histogram.
References

1. Galushkin A. I. Neyronnye seti. Osnovy teorii. [Neural networks. Fundamentals of the theory]. Moscow, Goryachaya liniya-Telekom Publ., 2010, 496 p.

2. Saymon Khaykin. Neyronnye seti. Polnyy kurs. [Neural networks. Full course]. Moscow, Vil'yams Publ., 2006, 1104 p.

3. GOST R 52633.5–2011. Zashchita informatsii. Tekhnika zashchity informatsii. Avtomaticheskoe obuchenie neyrosetevykh preobrazovateley biometriya-kod dostupa [Information protection. Information protection technology. The neural net biometry-code convertor automatic training]. Moscow, Standartinform Publ., 2012, 20 p.

4. Richard O. Duda, Peter E. Hart. Raspoznavanie obrazov i analiz stsen. [Pattern classification and scene analysis]. Moscow, Mir Publ., 1976, 507 p.

5. Ruud Bolle, Jonathan H. Connell, Sharanth Pankanti, Nalini K. Ratha, Andrew W. Senior. Rukovodstvo po biometrii [Guide to Biometrics]. Moscow, Technosphera Publ., 2007, 368 p.

6. Ivanov A. I. [Comparative analysis of competing technologies, biometrics cryptographic authentication identity]. Zashchita informatsii. Insaid. 2014, no. 3, p. 3–9. (In Russ.)

7. Efron B. Bootstrap methods: Another look at the Jackknife. Ann. Statistics, 7:1-26, 1979.

8. GOST R 52633.2–2010. Zashhita informacii. Tehnika zashhity informacii. Trebovanija k formirovaniju sinteticheskih biometricheskih obrazov, prednaznachennyh dlja testirovanija sredstv vysokonadezhnoj biometricheskoj autentifikacii. [Information protection. Information protection technology. Requirements for creation procedures for bases of synthetic biometric images, intended for high-reliability biometric authentication means testing]. Moscow, Standartinform Publ., 2011, 24 p.

9. Volchihin V. I., Ivanov A. I., Funtikov V. A., Nazarov I. G, Yazov Y. K. Nejrosetevaya zashhita personalnyh biometricheskix dannyh. [Neural network protection of personal biometric data]. Moscow, Radiotekhnika Publ., 2012., 160 p.

10. Akhmetov B. S., Volchikhin V. I., Ivanov A. I., Malygin A. Yu. Algoritmy testirovaniya biometriko-neyrosetevykh mekhanizmov zashchity informatsii. [Biometrics-testing algorithms neural network information protection mechanisms]. Kazakhstan, Almaty, KazNTU im. Satpaeva Publ., 2013, 152 p.

11. Akhmetov B. S., Ivanov A. I., Funtikov V. A., Bezyaev A. V., Malygina E. A. Tekhnologiya ispol'zovaniya bol'shikh neyronnykh setey dlya preobrazovaniya nechetkikh biometricheskikh dannykh v kod klyucha dostupa. [The technology of using large neural networks to convert the fuzzy biometric data in the key code access]. Kazakhstan, Almaty, TOO “Izdatel'stvo LEM” Publ., 2014, 144 p.

12. GOST R 52633.1–2009. Zashhita informacii. Tehnika zashhity informacii. Trebovanija k formirovaniju baz estestvennyh biometricheskih obrazov, prednaznachennyh dlja testirovanija sredstv vysokonadezhnoj biometricheskoj autentifikacii. [Information protection. Information protection technology. Requirements for creation procedures for bases of natural biometric images, intended for high-reliability biometric authentication means testing]. Moscow, Standartinform Publ., 2010, 24 p.

13. Ivanov A. I. Biometricheskaya identifikatsiya lichnosti po dinamike podsoznatel'nykh dvizheniy. [Biometric identification on the dynamics of the unconscious movements]. Penza, Penzenskiy gosudarstvennyy universitet Publ., 2000, 178 p.

14. Kulikov S. V., Sekretov M. V., Zakharov O. S., Ivanov A. I., Mayorov A. V. [Accounting for the "heavy" tails of the distribution of abnormal biometrics all “Aliens” when setting up the nonlinear element of a neuron with multiple discrete states]. Neyrokomp'yutery: razrabotka, primenenie, 2012, no. 3, p. 56–59 (In Russ.)

15. Volchikhin V. I., Ivanov A. I., Funtikov V. A., Malygina E. A. [Prospects for the use of artificial neural networks with multi-level quantizers in biometrics technology, neural network authentication]. Izvestiya vysshikh uchebnykh zavedeniy. Povolzhskiy region. Tekhnicheskie nauki. 2013, vol. 28, no. 4, p. 88–89.
(In Russ.)

 


Kachalin Sergey Viktorovich – postgraduate student, Penza State University. e-mail: s.kachalin@gmail.com