- Physical Review Journals
All JournalsPhysics Magazine

Physical Review Research

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Letter
  • Open Access

Crisis-induced intermittency in reservoir computing

Tempei Kabayama1,*, Motomasa Komuro2, Yasuo Kuniyoshi1, Kazuyuki Aihara2, and Kohei Nakajima1,2

  • 1Graduate School of Information Science and Technology, University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan
  • 2International Research Center for Neurointelligence, University of Tokyo Institutes for Advanced Study, University of Tokyo, Bunkyo-ku, Tokyo 113-0033, Japan

  • *Contact author: t-kabayama@isi.imi.i.u-tokyo.ac.jp
PDF

Phys. Rev. Research 7, L032058 – Published 12 September, 2025

DOI: https://doi.org/10.1103/vqvp-mbxx

Abstract

Reservoir computing enables embedding attractors into random neural networks (RNNs), often generating a “mirror” of a target attractor due to inherent symmetrical constraints in their architecture, which results in multistability. In this work, we report that a simple modulation of a global parameter—the scale of internal connectivity—can induce an attractor-merging crisis accompanied by intermittency, which is characterized by irregular alternation between staying within and escaping from the ruins of embedded attractors. We further reveal its underlying mechanism through a detailed analysis of the phase-space structure, and multiple case studies demonstrate that this bifurcation scenario is intrinsic to a general class of RNNs, independent of training data. This work bridges the gap between machine learning implementations and concepts in nonlinear dynamical systems, contributing to the design of systems that leverage the intrinsic properties of RNNs, as well as to a deeper understanding of biological neural dynamics.

View figure in article

Physics Subject Headings (PhySH)

Article Text

Supplemental Material

References (53)

  1. K. Aihara and G. Matsumoto, Forced oscillations and routes to chaos in the Hodgkin-Huxley axons and squid giant axons, in Chaos in Biological Systems, edited by H. Degn, A. V. Holden, and L. F. Olsen (Springer US, Boston, MA, 1987), pp. 121–131.
  2. W. J. Freeman, Simulation of chaotic EEG patterns with a dynamic model of the olfactory system, Biol. Cybern. 56, 139 (1987).
  3. K. Inoue, S. Ohara, Y. Kuniyoshi, and K. Nakajima, Transient chaos in bidirectional encoder representations from transformers, Phys. Rev. Res. 4, 013204 (2022).
  4. L. Storm, H. Linander, J. Bec, K. Gustavsson, and B. Mehlig, Finite-time Lyapunov exponents of deep neural networks, Phys. Rev. Lett. 132, 057301 (2024).
  5. I. Tokuda, T. Nagashima, and K. Aihara, Global bifurcation structure of chaotic neural networks and its application to traveling salesman problems, Neural Netw. 10, 1673 (1997).
  6. S. Liu, N. Akashi, Q. Huang, Y. Kuniyoshi, and K. Nakajima, Exploiting chaotic dynamics as deep neural networks, Phys. Rev. Res. 7, 033031 (2025).
  7. K. Ikeda, K. Otsuka, and K. Matsumoto, Maxwell-Bloch turbulence, Prog. Theor. Phys. Suppl. 99, 295 (1989).
  8. K. Kaneko, Clustering, coding, switching, hierarchical ordering, and control in a network of chaotic elements, Physica D 41, 137 (1990).
  9. I. Tsuda, Chaotic itinerancy as a dynamical basis of hermeneutics in brain and mind, World Futures 32, 167 (1991).
  10. R. Iwami, T. Mihana, K. Kanno, S. Sunada, M. Naruse, and A. Uchida, Controlling chaotic itinerancy in laser dynamics for reinforcement learning, Sci. Adv. 8, eabn8325 (2022).
  11. K. Inoue, K. Nakajima, and Y. Kuniyoshi, Designing spontaneous behavioral switching via chaotic itinerancy, Sci. Adv. 6, eabb3989 (2020).
  12. H. Jaeger, The “echo state” approach to analysing and training recurrent neural networks—with an erratum note, GMD Technical Report No. 148 (German National Research Center for Information Technology, Bonn, Germany, 2001), p. 13.
  13. W. Maass, T. Natschläger, and H. Markram, Real-time computing without stable states: A new framework for neural computation based on perturbations, Neural Comput. 14, 2531 (2002).
  14. In Reservoir Computing: Theory, Physical Implementations, and Applications, edited by K. Nakajima and I. Fischer (Springer, Singapore, 2021).
  15. Z. Lu, B. R. Hunt, and E. Ott, Attractor reconstruction by machine learning, Chaos 28, 061104 (2018).
  16. A. Hart, J. Hook, and J. Dawes, Embedding and approximation theorems for echo state networks, Neural Netw. 128, 234 (2020).
  17. J. Z. Kim, Z. Lu, E. Nozari, G. J. Pappas, and D. S. Bassett, Teaching recurrent neural networks to infer global temporal structure from local examples, Nat. Mach. Intell. 3, 316 (2021).
  18. A. Flynn, V. A. Tsachouridis, and A. Amann, Multifunctionality in a reservoir computer, Chaos 31, 013125 (2021).
  19. A. Flynn, V. A. Tsachouridis, and A. Amann, Seeing double with a multifunctional reservoir computer, Chaos 33, 113115 (2023).
  20. A. Flynn, Theory and applications of multifunctional reservoir computers, Ph.D. dissertation, University College Cork, Cork, Ireland, 2023.
  21. C. Grebogi, E. Ott, and J. A. Yorke, Chaotic attractors in crisis, Phys. Rev. Lett. 48, 1507 (1982).
  22. C. Grebogi, E. Ott, F. Romeiras, and J. A. Yorke, Critical exponents for crisis-induced intermittency, Phys. Rev. A 36, 5365 (1987).
  23. C. Grebogi, E. Ott, and J. A. Yorke, Crises, sudden changes in chaotic attractors, and transient chaos, Physica D 7, 181 (1983).
  24. J. Namikawa, Chaotic itinerancy and power-law residence time distribution in stochastic dynamical systems, Phys. Rev. E 72, 026204 (2005).
  25. M. Komuro, A mechanism of chaotic itinerancy in globally coupled maps (singular phenomena of dynamical systems), RIMS Kôkyûroku 1118, 96 (1999) (in Japanese).
  26. G. Tanaka, M. A. F. Sanjuán, and K. Aihara, Crisis-induced intermittency in two coupled chaotic maps: Towards understanding chaotic itinerancy, Phys. Rev. E 71, 016219 (2005).
  27. L.-W. Kong, H.-W. Fan, C. Grebogi, and Y.-C. Lai, Emergence of transient chaos and intermittency in machine learning, J. Phys.: Complexity 2, 035014 (2021).
  28. L.-W. Kong, H.-W. Fan, C. Grebogi, and Y.-C. Lai, Machine learning prediction of critical transition and system collapse, Phys. Rev. Res. 3, 013090 (2021).
  29. A. Flynn and A. Amann, Exploring the origins of switching dynamics in a multifunctional reservoir computer, Front. Netw. Physiol. 4, 1451812 (2024).
  30. H. Jaeger, M. Lukoševičius, D. Popovici, and U. Siewert, Optimization and applications of echo state networks with leaky-integrator neurons, Neural Netw. 20, 335 (2007).
  31. H. Sompolinsky, A. Crisanti, and H. J. Sommers, Chaos in random neural networks, Phys. Rev. Lett. 61, 259 (1988).
  32. B. Cessac and M. Samuelides, From neuron to neural networks dynamics, Eur. Phys. J.: Spec. Top. 142, 7 (2007).
  33. T. Haruna and K. Nakajima, Optimal short-term memory before the edge of chaos in driven random recurrent networks, Phys. Rev. E 100, 062312 (2019).
  34. D. Verstraeten, B. Schrauwen, M. d'Haene, and D. Stroobandt, An experimental unification of reservoir computing methods, Neural Netw. 20, 391 (2007).
  35. See Supplemental Material at http://link.aps.org/supplemental/10.1103/vqvp-mbxx for details of methods, additional analyses, and corresponding videos.
  36. E. N. Lorenz, Deterministic nonperiodic flow, J. Atmos. Sci. 20, 130 (1963).
  37. O. I. Moskalenko, M. K. Kurovskaya, and A. A. Koronovskii, Intermittent generalized synchronization in unidirectionally coupled systems with different topology of attractors, Eur. Phys. J.: Spec. Top. 233, 1 (2024).
  38. A. E. Hramov and A. A. Koronovskii, Intermittent generalized synchronization in unidirectionally coupled chaotic oscillators, Europhys. Lett. 70, 169 (2005).
  39. L. Zhao, Y.-C. Lai, and C.-W. Shih, Transition to intermittent chaotic synchronization, Phys. Rev. E 72, 036212 (2005).
  40. J. Guckenheimer and R. F. Williams, Structural stability of Lorenz attractors, Publ. Math. l'IHÉS 50, 59 (1979).
  41. T. Kabayama, Y. Kuniyoshi, K. Aihara, and K. Nakajima, Designing chaotic attractors: A semisupervised approach, Phys. Rev. E 111, 034207 (2025).
  42. L. Chua, M. Komuro, and T. Matsumoto, The double scroll family, IEEE Trans. Circuit Syst. 33, 1072 (1986).
  43. A. Sordi, Chua’s oscillator: an introductory approach to chaos theory, Rev. Bras. Ensino Fís. 43, e20200437 (2020).
  44. R. Laje and D. V. Buonomano, Robust timing and motor patterns by taming chaos in recurrent neural networks, Nat. Neurosci. 16, 925 (2013).
  45. J. Namikawa, R. Nishimoto, and J. Tani, A neurodynamic account of spontaneous behaviour, PLoS Comput. Biol. 7, e1002221 (2011).
  46. S. Steingrube, M. Timme, F. Wörgötter, and P. Manoonpong, Self-organized adaptation of a simple neural circuit enables complex robot behaviour, Nat. Phys. 6, 224 (2010).
  47. J. Schulman, P. Moritz, S. Levine, M. Jordan, and P. Abbeel, High-dimensional continuous control using generalized advantage estimation, arXiv:1506.02438.
  48. E. Coumans and Y. Bai, Pybullet, a python module for physics simulation for games, robotics and machine learning, 2016–2021, http://pybullet.org.
  49. K. Nakajima, Physical reservoir computing—An introductory perspective, Jpn. J. Appl. Phys. 59, 060501 (2020).
  50. N. Akashi, Y. Kuniyoshi, T. Jo, M. Nishida, R. Sakurai, Y. Wakao, and K. Nakajima, bifurcations into pneumatic artificial muscle, Adv. Sci. 11, 2304402 (2024).
  51. F. Hadaeghi, M. R. Hashemi Golpayegani, and K. Moradi, Does “crisis-induced intermittency” explain bipolar disorder dynamics? Front. Comput. Neurosci. 7, 116 (2013).
  52. F. Hadaeghi, M. R. H. Golpayegani, and G. Murray, Towards a complex system understanding of bipolar disorder: A map based model of a complex winnerless competition, J. Theor. Biol. 376, 74 (2015).
  53. S. Yonekura, A. Narita, H. Kanazawa, and Y. Kuniyoshi, A reservoir model of respiration-induced perceptual alternation in binocular rivalry, bioRxiv (2025), doi: 10.1101/2025.02.06.636993.

Outline

Information

Phys. Rev. Research 7, L032058– Published 12 September, 2025
  • Received 17 April 2025
  • Accepted 26 August 2025
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation