Thermodynamic Formalism for the Study and Formation of Algorithms and Neural Networks

Thermodynamic Formalism for the Study and Formation of Algorithms and Neural Networks

I.F. Yasinskiy | F.N. Yasinskiy

Ivanovo State Textile Academy, Russia

Ivanovo State Power University, Russia

| |
| | Citation



This article considers the questions connected with creation of optimum algorithms using the laws of thermodynamics as applied to a computing process. Ideas and methods of phenomenological and statistic thermodynamics are used to estimate the amount of calculations or volume of the neural network. Introduction of the other thermodynamic functions, besides entropy, and also definition of the three thermodynamic origins in the context of calculations allow to study stability, organize the parameters according their information weights, carry out the decomposition of the complex systems, construct the rapid algorithms. The way of creation of the neural network structure is offered, consisting in use of the pre-trained fragments.


Computational entropy, neural network structure, rapid algorithms, thermodynamics of calculations


[1] Hinchin, A.Y., The concept of entropy in probability theory, UMN, tome 11, 3(55), 1953.

[2] Kolmogorov, A.N. & Tihomirov, V.M., e -entropy and e -capacity of sets in functional spaces, UMN, tome 14, 2(86), pp. 3–80, 1959.

[3] Vitushkin, A.G., Assessment of the Tabulation Problem, Fizmatiz: Moscow, 1959.

[4] Petri, N.V., Complexity of the algorithms and time of the their work, DAN USSR, pp. 30–31, 1969.

[5] Barak, B., Shaltiel, R. & Wigderson, A., Computational analogues of entropy. Proc. of the 11th International Conference on Random Structures and Algorithms, pp. 200–215, Springer, 2003.

[6] Haitner, I., Reingold, O., Vadhan, S. & Wee, H., Inaccessible entropy. Proc. of the 41st Annual ACM Symposium on Theory of Computing (STOC `09), pp. 611–620, 31 May–2 June, 2009.

[7] Kolmogorov, A.N., The different approaches to the diffi culty estimation of the approximate setting and computing of functions. Proc. of the Int. Congress of Mathematicians, Stockholm, 1964.

[8] Babenko, K.I., Theoretical Foundations and Construction of Numerical Algorithms for Problems of Mathematical Physics, Nauka: Moscow, p. 295, 1979.

[9] Aho, A., Hopkroft, J. & Ulman, J., The Design and Analysis of Computational Algorithms, Mir: Moscow, p. 535, 1979.

[10] Traub, J. & Voznyakovskiy, H., The General Theory of Optimal Algorithms, Mir: Moscow, p. 381, 1983.

[11] Fuller, B. & Reyzin L., Computational Entropy and Information Leakage (2011), available at

[12] Vatolin, U.N., Information criterion of stability. Numerical Methods of Mechanics of the Continuous Environment, 2(3), 1971.

[13] Vatolin, U.N., On the application of the entropy bounds of stability, numerical methods of continuum mechanics. Numerical Methods of Mechanics of the Continuous Environment, 5(2), pp. 5–6, 1974.

[14] Haykin, S., Neural Networks, Williams: Moscow, pp. 89–340, 2006.

[15] Kovalevskiy, S.V. & Reshetnik, N.A., Features of systems diagnostics with application of a principle of an entropy minimum. Proc. of the VII All-Russia Conf. On Neuro computers and their application, Moscow, pp. 394–396, 2001.