An approach to self-training of a high-order autoassociative neural network model

Purpose of reseach. To develop a self-learning algorithm for a two-layer neural network model that is as efficient as possible, given the current technical implementation of intelligent systems, including those designed for solving pattern recognition problems. This algorithm will be based on increasing the number of neurons and varying the weight coefficients of synaptic connections, with the possibility of extending it to a high-order multiconnected neural network with an internal product of vectors.

Methods. To solve this problem, this paper proposes an approach to synthesizing a high-order multiconnected neural network model with an internal product of vectors, as well as a self-learning algorithm for such a neural network. This algorithm provides for the rapid correction of the elements of the reference matrix, instead of the traditional variation of the weight coefficients of synaptic connections, in order to reduce the resource intensity of the performed operations.

Results. The proposed method was implemented as a software application linked to the self-training of a high-order neural network using speech sound types represented in raster format, pre-segmented from the general stream and transformed into polar coordinates for ease of processing and storing the resulting images as a training set.

Conclusion. The developed algorithm, during software implementation testing, demonstrated relatively high efficiency by eliminating the resource-intensive operation of varying weight coefficients and replacing it with direct correction of the reference matrix. This algorithm demonstrated relatively high efficiency, convergence in a finite number of steps due to the limited number of first-approximation codes of reference vectors, and noticeable performance compared to known analogs.

1. Bobyr M., Kulabukhov S., Milostnaya N. Training a neuro-fuzzy system based on the area difference method. Iskusstvennyi intellect i prinyatie resheniy= Artificial Intelligence and Decision Making. 2016; (4): 15-26. (In Russ.).

2. Schmidhuber J. Deep learning in neural networks: an overview. Neural networks. 2015; 61: 85-117. (In Russ.).

3. Emelianov S., Bobyr M., Antsiferov A. Self-learning algorithm for adaptive neurofuzzy systems based on soft computing. Informatsionno-izmeritelnye i upravlyaushie sistemy = Information, measuring and control systems. 2013; (8): 3-9. (In Russ.).

4. Malyshev A., Tipikin A., Taranenko K. Self-training of an autoassociative model of a high-order neural network. Izvestiya Kurskogo gosudarstvennogo tekhnicheskogo universiteta = Proceeding of Kursk State Technical University. 1998; (2): 63-68. (In Russ.).

5. Ashwin P., Coombes S., Nicks R. Mathematical frameworks for oscillatory network dynamics in neuroscience. Journal of mathematical neuroscience. 2016; 6: 1-92.

6. Malyshev A., Novikova A. Neural algorithm of reconfiguration for matrix multiprocessor. In: International Conference «Modern trends in education and science». Tambov; 2014. P. 10-11.

7. Bobyr M. Training of neuro-fuzzy systems. Moscow: Argamak-Media; 2017. 240 p. (In Russ.).

8. Chen C., Chen J., Chen M., Bao B., Bao H. Coexisting multi-stable patterns in memristor synapse-coupled Hopfield neural network with two neurons. Nonlinear dynamics. 2019; 95: 3385-3399.

9. Malyshev A., Alekseev A., Gevorkyan A. Recognition and tracking of objects using neural network model tools. In: Programmnaya inzheneriya: sovremennye tendentsii razvitiya i primeneniya: materialy VIII Vserossiiskoi nauchno-prakticheskoi konferentsii = Proceedings of the VIII All-Russian scientific and practical conference «Software engineering: modern trends in development and application». Kursk; 2024. P. 97-101. (In Russ.).

10. Lai Q., Wan ZH., Zhang H., Chen G. Design and analysis of multiscroll memristive Hopfield neural network with adjustable memductance and application to image encryption. IEEE Transactions on neural networks and learning systems. 2023; 34: 7824-7837.

11. Danca M., Kuznetsov N. Hidden chaotic sets in a Hopfield neural system. Chaos, solitons & fractals. 2017; 103: 144-150.

12. Rebentrost P., Bromley T., Weedbrook C., Lloyd S. Quantum Hopfield neural network. Physical review. 2018; 98: 042308.

13. Malyshev A., Glukhov N. Application of artificial intelligence for the analysis of the life cycle of a technological project. In: Molodezh' i XXI vek: materialy 13-i Mezhdunarodnoi nauchnoi konferentsii = Proceedings of the 13th International Scientific Conference «Youth and the 21st Century». Kursk; 2024. P. 348-351. (In Russ.).

14. Ong B., Sugiura K., Zettsu K. Dynamically pre-trained deep recurrent neural networks using environmental monitoring data for predicting. Neural computing & applications. 2016; 27: 1553-1566.

15. Malyshev A., Krylov P., Volodin O. Metaheuristic analytical instruments in optimizing the shopping center parking area: a transportation model. International journal of advanced science and technology. 2019; 28: 198-207.

16. Malyshev A.V., Puzyrev E.I., Prokhorov M.V., Nefedov N.G. The Program of the Cardiovascular Disease Risk Prediction System. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel'naya tekhnika, informatika. Meditsinskoe priborostroenie = Proceedings of the Southwest State University. Series: Control, Computing Engineering, Information Science. Medical Instruments Engineering. 2022;12(3):46-61. (In Russ.). https://doi.org/10.21869/2223-1536-2022-12-3-46-61

17. Dzyubin I., Malyshev A. Methods for optimizing highly loaded applications. In: Materialy V Vserossiiskoi nauchno-prakticheskoi konferentsii «Programmnaya inzheneriya: sovremennye tendentsii razvitiya i primeneniya» = Proceedings of the V All-Russian scientific and practical conference «Software engineering: modern trends in development and application». Kursk; 2021. P. 12-20. (In Russ.).

18. Chaplygin A., Malyshev A. An operating system with a kernel based on a virtual machine with a reduced instruction set. Izvestiya Yugo-Zapadnogo gosudarstvennogo universiteta. Seriya: Upravlenie, vychislitel'naya tekhnika, informatika. Meditsinskoe priborostroenie = Proceedings of the Southwest State University. Series: Control, Computing Engineering, Information Science. Medical Instruments Engineering, 2020; 10(1): 8-20. (In Russ.).

19. Malyshev A., Rodin A., Saburov V. Design of an automated translation system for mobile devices. In: Materialy Vserossiiskoi nauchnoi konferentsii «Programmnaya inzheneriya: sovremennye tendentsii razvitiya i primeneniya» = Proceedings of the All-Russian scientific and practical conference «Software engineering: modern trends in development and application». Kursk; 2017. P. 143-147. (In Russ.).

20. Vorobyova I., Psareva A., Malyshev A. Methods of semantic clustering of texts. In: Materialy V Vserossiiskoi nauchno-prakticheskoi konferentsii «Programmnaya inzheneriya: sovremennye tendentsii razvitiya i primeneniya» = Proceedings of the V All-Russian scientific and practical conference «Software engineering: modern trends in development and application». Kursk; 2021. P. 81-84. (In Russ.).