Wave regression: nonlinear cognitive heuristic

The quality of regression is determined by the choice of an approximation function, more or less accurately reflecting the process which generated the data. An important class of such processes is cognitive processes of largely wave nature. Here, the corresponding wave-like calculus is used in the new method of behavioral regression. We generalize classical linear regression from real weights to complex-valued amplitudes the modules and phases of which encode the amplification and delay of cognitive waves. The target feature then emerges as squared module of total amplitude influences of all basis features. The obtained regression models are tested on the data of academic performance of the study group in comparison with linear regressions of the same number of parameters. When using all basis features, the accuracy of wave regression is close to the accuracy of linear models. With fewer basis features the quality of linear regression degrades, while the performance of wave regression improves. The largest difference is observed in triadic regime when the target feature is produced by two basis features. In this case, the error of three-parameter wave regression is 2.5 % lower than that of full linear regression with 21 parameters. This dramatic improvement is due to a special nonlinearity of wave regression, typical to pragmatic heuristics of natural thinking. This nonlinearity takes advantage of semantic correlations of features missed by classical regressions. The wave-like reduction of computational complexity opens up ways for developing more efficient and nature-like algorithms of data analysis and artificial intelligence.

1. Polezhaev V.D., Polezhaeva L.N. Nonlinear paired regression models in the econometrics course. Modern problems of science and education, 2018, no. 4, pp. 73. (in Russian)

2. Kolenteev N.Ia., Goncharova O.A., Goncharov V.S. Paired nonlinear regression and correlation. Proc. of the Innovative Technologies in Higher Education Pedagogy, 2022, pp. 304–312. (in Russian)

3. Khrennikov A. Quantum-like modeling of cognition. Frontiers in Physics, 2015, vol. 3, pp. 77. https://doi.org/10.3389/fphy.2015.00077

4. Asano M., Basieva I., Khrennikov A., Ohya M., Tanaka Y., Yamato I. Quantum information biology: from information interpretation of quantum mechanics to applications in molecular biology and cognitive psychology. Foundations of Physics, 2015, vol. 45, no. 10, pp. 1362–1378. https://doi.org/10.1007/s10701-015-9929-y

5. Surov I.A., Alodzhantc A.P. Decision Models in Quantum Cognitive Science. St. Petersburg, ITMO University, 2018. 63 p. (in Russian)

6. Surov I.A. Logic of sets and logic of waves in cognitive-behavioral modeling. Information and Mathematical Technologies in Science and Management, 2023, no. 4(32), pp. 51–66. https://doi.org/10.25729/ESI.2023.32.4.005 (in Russian)

7. Khrennikov A.Y. Ubiquitous Quantum Structure: From Psychology to Finance. Springer, 2010, 216 p. https://doi.org/10.1007/978-3-642-05101-2

8. Surov I.A., Pilkevich S.V., Alodjants, A.P., Khmelevsky S.V. Quantum phase stability in human cognition. Frontiers in Psychology, 2019, vol. 10, pp. 929. https://doi.org/10.3389/fpsyg.2019.00929

9. Shan Z.H. Brainwave phase stability: predictive modeling of irrational decision. Frontiers in Psychology, 2022, vol. 13, pp. 617051. https:// doi.org/10.3389/fpsyg.2022.617051

10. Head J.D., Zerner M.C. A Broyden—Fletcher—Goldfarb—Shanno optimization procedure for molecular geometries. Chemical Physics Letters, 1985, vol. 122, no. 3, pp. 264–270. https://doi.org/10.1016/0009-2614(85)80574-1

11. Oppenheim A.V., Lim J.S. The importance of phase in signals. Proceedings of the IEEE, 1981, vol. 69, no. 5, pp. 529–541. https://doi.org/10.1109/PROC.1981.12022

12. Sorkin R.D. Quantum mechanics as quantum measure theory. Modern Physics Letters A, 1994, vol. 9, no. 33, pp. 3119–3127. https://doi.org/10.1142/s021773239400294x

13. Bazilevskiy M.P. Nonlinear criteria of quasilinear regression models. Modeling, Optimization and Information Technology, 2018, vol. 6, no. 4 (23), pp. 185–195. (in Russian). https://doi.org/10.26102/2310-6018/2018.23.4.015

14. Surov I.A. Quantum cognitive triad: semantic geometry of context representation. Foundations of Science, 2021, vol. 26, no. 4, pp. 947–975. https://doi.org/10.1007/s10699-020-09712-x

15. Rudoi G.I. Inductive generation of superpositions in nonlinear regression problems. Mashinnoe obuchenie i analiz dannyh, 2011, vol. 1, no. 2, pp. 183–197. (in Russian)

16. Sologub R.A. Methods of the nonlinear regression model transformation. Mashinnoe obuchenie i analiz dannyh, 2015, vol. 1, no. 14, pp. 1961–1976. (in Russian)

17. Shestopal O.V., Chernoivan D.N., Seredina P.B. Robust methods of building and improving multidimensional linear and nonlinear regressions. T-Comm, 2019, vol. 13, no. 2, pp. 46–51. (in Russian). https://doi.org/10.24411/2072-8735-2018-10235

18. Marsh B., Todd P.M., Gigerenzer G. Cognitive heuristics. Reasoning the Fast and Frugal Way, 2003, pp. 273–288. https://doi.org/10.1017/cbo9780511818714.010

19. Strizhov V.V. Search for a model of optimal complexity in nonlinear regression problems. MMRO, 2005, vol. 12, no. 1, pp. 206–209. (in Russian)

20. Sologub R.A. Nonlinear model generation algorithms. Information Technologies, 2013, no. 5, pp. 8–12. (in Russian)

21. Shiriaev V.D., Shagilova E.V., Beliakov M.T. About the form choice of the nonlinear regression model. Proc. of the Youth and Systemic Modernization of the Country, 2024, pp. 137–142. (in Russian)

22. Chen S. How much energy will Al really consume? The good, the bad and the unknown. Nature, 2025, vol. 639, no. 8053, pp. 22–24. https://doi.org/10.1038/d41586-025-00616-z

23. Kureichik V.V., Rodzin S.I. Bio-heuristics inspired by fauna (review). Information Technologies, 2023, vol. 29, no. 11, pp. 559–573. (in Russian). https://doi.org/10.17587/it.29.559-573

24. Surov I.A. Process semantics of complex numbers. Mathematical Structures and Modeling, 2023, vol. 68, no. 4, pp. 71-84. (in Russian). https://doi.org/10.24147/2222-8772.2023.4.71-84

25. Surov I.A. What is the difference? Pragmatic formalization of meaning. Artificial Intelligence and Decision Making, 2023, no. 1, pp. 78–89. (in Russian). https://doi.org/10.14357/20718594230108

26. Surov I.A. Сolor coding of qubit states. Informatics and Automation, 2023, vol. 22, no. 5, pp. 1207–1236. (in Russian). https://doi.org/10.15622/ia.22.5.9

27. Pavlov A.V. Implementation of regression models of data processing by the fourier-holography method. Journal of Computer and Systems Sciences International, 2005, vol. 44, no. 2, pp. 184–191.

28. Lin X., Rivenson Y., Yardimei N.T., Veli M., Luo Y., Jarrahi M., Ozcan A. All-optical machine learning using diffractive deep neural networks. Science, 2018, vol. 361, no. 6406, pp. 1004–1008. https://doi.org/10.1126/science.aat8084

29. Krasnov A.E., Golovkin M.E., Nikolskii D.N., Blagoveshchenskii V.G. Wave network for image recognition. Avtomatizacija v Promyshlennosti, 2022, no. 10, pp. 28–33. (in Russian). https://doi.org/10.25728/avtprom.2022.10.06

30. Raikov A.N. All-analogue photonic artificial intelligence. Information Society, 2024, no. 6, pp. 168–179. (in Russian)