pith. sign in

arxiv: 2312.12264 · v2 · pith:CWZS2TTTnew · submitted 2023-12-19 · ⚛️ physics.chem-ph · cond-mat.dis-nn

Exploring Non-Steady-State Charge Transport Dynamics in Information Processing: Insights from Reservoir Computing

Zheyang Li , Xi Yu This is my paper

Pith reviewed 2026-05-24 05:37 UTC · model grok-4.3

classification ⚛️ physics.chem-ph cond-mat.dis-nn
keywords reservoir computingnon-steady-state charge transportproton-coupled dynamicsinformation processingband-pass filternonlinear chemical systemsmemory capacityin-materia computing
0
0 comments X

The pith

The alignment between RC signal processing frequency and the characteristic time of a nonlinear chemical system's dynamics dictates task efficiency, reservoir states, and memory capacity.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper integrates a non-steady-state proton-coupled charge transport system into a reservoir computing architecture for information processing. It reports successful performance on waveform recognition, voice identification, and chaos system prediction. The central finding is that matching the RC processing frequency to the system's characteristic dynamic time scale controls how well tasks execute, what states the reservoir reaches, and how much memory capacity is available. This match also sets the range of frequencies the system can handle, producing behavior like a chemically tuned band-pass filter that selects specific frequencies. The work positions such dynamic molecular systems as a route to in-materia computing.

Core claim

By embedding a non-steady-state proton-coupled charge transport system into reservoir computing, the architecture performs waveform recognition, voice identification, and chaos prediction. Efficiency of these tasks, the reservoir states reached, and memory capacity are all governed by how closely the RC signal processing frequency aligns with the characteristic time of the nonlinear dynamics. The characteristic time further tunes the usable frequency window, yielding an effect equivalent to a chemically tuned band-pass filter for selective frequency processing.

What carries the argument

Non-steady-state proton-coupled charge transport system placed inside reservoir computing, where frequency alignment with the system's characteristic time sets processing performance and creates selective frequency response.

If this is right

  • Waveform recognition, voice identification, and chaos prediction become feasible when frequencies are matched.
  • Reservoir states and memory capacity vary directly with the degree of frequency alignment.
  • The usable information processing frequency range is set by the dynamic system's characteristic time.
  • The overall behavior functions as a chemically tuned band-pass filter for selective frequency handling.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same frequency-matching rule may apply to other nonlinear chemical or physical systems used for computing.
  • Physical device implementations could test whether the predicted band-pass filtering appears in measured signals.
  • Frequency-selective molecular processors might be designed for tasks that require isolating particular signal bands.

Load-bearing premise

The theoretical model of the proton-coupled charge transport system correctly describes its real dynamics once placed inside the reservoir computing framework.

What would settle it

Measurements in which the RC frequency is swept across the system's characteristic time while task accuracy, reservoir state distribution, and memory capacity show no corresponding peak or modulation.

Figures

Figures reproduced from arXiv: 2312.12264 by Xi Yu, Zheyang Li.

Figure 1
Figure 1. Figure 1: the mechanism and characteristics of the non-steady-state charge transport dynamics (a) The scheme of PCET dynamics (The water in the environment act as the proton source here). Electron transport through ￾￾ or ￾￾ − ￾￾￾ channel modulate the occupation of redox state, resulting in the proton transfer reaction and channels conversion (b) A general redox reaction coupling with electron transport dynamics. The… view at source ↗
Figure 2
Figure 2. Figure 2: Schematic of the non-steady-state reservoir computing system.(a) The input signal ￾(￾) is multiplexed through a mask before entering the cyclic reservoir with ￾ virtual nodes. The output weight ￾￾￾￾ is trained using linear regression and functions as the weight matrix to map the reservoir states to the output. (b) The masking procedure is as follows: an input signal is sampled over a duration ￾ and then mu… view at source ↗
Figure 4
Figure 4. Figure 4: Prediction tasks demonstrations by the non-steady-state dynamic RC (a) Hénon map task: The predicted values of ￾(￾) and a two-dimensional plot of the intended targets (in blue) with the RC-generated outputs (in red). The results indicate that a temporal interval ￾ of 0.05 ￾ yields suboptimal predictions when compared to an interval ￾ of 2.5 ￾ . (b) Lorenz atmospheric task: The predicted results and a three… view at source ↗
Figure 5
Figure 5. Figure 5: Quantitative analysis of RC performance and dynamic characteristics in the non-steady-state system. (a) Normalized successful recognition counts for 44 waveform recognition tasks on a logarithmic scale and peak at ￾ = 10￾Hz. (b) Five reservoir states observed during the waveform recognition task. The raw waveform is represented by a gray line, and an inset panel provides a detailed view of a specific state… view at source ↗
Figure 7
Figure 7. Figure 7: Humidity dependent experiments of the non-steady-state dynamic RC. (a) Normalized counts of successful waveform recognition across varying input frequencies. Environmental water concentration dynamically modulates peak shifts in recognition performance from 0.1 to 100% relative humidity, maintaining a consistent trend. (b) Schematic of the environmentally controlled non-steady-state dynamic device for neur… view at source ↗
read the original abstract

Exploring nonlinear chemical dynamic systems for information processing has emerged as a frontier in chemical and computational research, seeking to replicate the brain's neuromorphic and dynamic functionalities. We have extensively explored the information processing capabilities of a nonlinear chemical dynamic system through theoretical modeling by integrating a non-steady-state proton-coupled charge transport system into reservoir computing (RC) architecture. Our system demonstrated remarkable success in tasks such as waveform recognition, voice identification and chaos system prediction. More importantly, through a quantitative study, we revealed the key role of the alignment between the signal processing frequency of the RC and the characteristic time of the dynamics of the nonlinear system, which dictates the efficiency of RC task execution, the reservoir states and the memory capacity in information processing. The system's information processing frequency range was further modulated by the characteristic time of the dynamic system, resulting in an implementation akin to a 'chemically-tuned band-pass filter' for selective frequency processing. Our study thus elucidates the fundamental requirements and dynamic underpinnings of the non-steady-state charge transport dynamic system for RC, laying a foundational groundwork for the application of dynamic molecular devices for in-materia computing.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript presents a theoretical modeling study that integrates a non-steady-state proton-coupled charge transport system into a reservoir computing (RC) architecture. It claims success on waveform recognition, voice identification, and chaos prediction tasks, and reports that alignment between the RC signal-processing frequency and the characteristic dynamical time of the chemical system controls task efficiency, reservoir states, and memory capacity. The work further asserts that the characteristic time modulates the usable frequency range, producing behavior analogous to a chemically tuned band-pass filter for selective frequency processing.

Significance. If the quantitative dependence of performance metrics on frequency alignment is robustly demonstrated and the model is shown to capture the essential dynamics, the result would supply a concrete design principle for using non-steady-state chemical systems in physical reservoir computing. The band-pass-filter analogy, if supported by explicit calculations of memory capacity versus input frequency, would be a useful conceptual contribution to in-materia computing.

major comments (2)
  1. [Abstract] Abstract: the central claims of 'remarkable success' on the listed tasks and of a 'key role' for frequency alignment are stated without any numerical performance metrics, error analysis, baseline comparisons, or figures. Because these quantitative results are required to substantiate the frequency-alignment mechanism, their absence prevents evaluation of the primary claim.
  2. [Abstract] Abstract and modeling section: the characteristic time of the proton-coupled charge transport dynamics is invoked as the quantity that sets the band-pass window, yet no rate equations, parameter values, or derivation of this time scale are supplied. Without these, it is impossible to assess whether unmodeled effects (additional reaction channels, electrode kinetics, or solvent relaxation) would shift the alignment condition that is asserted to dictate performance.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments on our manuscript. We address each major comment below and indicate the revisions made to strengthen the presentation of our results.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claims of 'remarkable success' on the listed tasks and of a 'key role' for frequency alignment are stated without any numerical performance metrics, error analysis, baseline comparisons, or figures. Because these quantitative results are required to substantiate the frequency-alignment mechanism, their absence prevents evaluation of the primary claim.

    Authors: We agree that the abstract should include quantitative metrics to support the claims. The full manuscript contains performance metrics, error bars, baseline comparisons, and figures demonstrating the frequency-alignment effects on task accuracy, reservoir states, and memory capacity. In the revised version we have updated the abstract to report key numerical results (e.g., task accuracies and memory-capacity values) together with explicit references to the relevant figures. revision: yes

  2. Referee: [Abstract] Abstract and modeling section: the characteristic time of the proton-coupled charge transport dynamics is invoked as the quantity that sets the band-pass window, yet no rate equations, parameter values, or derivation of this time scale are supplied. Without these, it is impossible to assess whether unmodeled effects (additional reaction channels, electrode kinetics, or solvent relaxation) would shift the alignment condition that is asserted to dictate performance.

    Authors: We acknowledge that the abstract and modeling section should make the derivation of the characteristic time fully explicit. The manuscript derives this time scale from the underlying proton-coupled charge-transport kinetics; to address the concern we have expanded both the abstract and the modeling section in the revision to include the explicit rate equations, the numerical parameter values employed, and the step-by-step derivation of the characteristic time. This addition allows direct evaluation of possible unmodeled effects. revision: yes

Circularity Check

0 steps flagged

No circularity detected; derivation chain not visible in text

full rationale

The provided abstract and context contain no equations, derivations, fitted parameters, or self-citations that could be inspected for reduction to inputs by construction. The central claims about frequency alignment and band-pass behavior are presented as outcomes of a quantitative study, but without any load-bearing steps, ansatzes, or uniqueness theorems shown, the derivation remains self-contained against external benchmarks. No patterns matching self-definitional, fitted-input, or self-citation circularity are identifiable.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no information on free parameters, axioms, or invented entities.

pith-pipeline@v0.9.0 · 5729 in / 1144 out tokens · 30666 ms · 2026-05-24T05:37:35.407531+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

35 extracted references · 35 canonical work pages

  1. [1]

    (1) Hunt, K.; Hunt, P.; Ross, J., Nonlinear dynamics and thermodynamics of chemical reactions far from equilibrium. Annu. Rev. Phys. Chem. 1990, 41 (1), 409 -

    work page 1990

  2. [2]

    (2) Sagués, F.; Epstein, I

    https://doi.org/10.1146/annur ev.pc.41.100190.002205 . (2) Sagués, F.; Epstein, I. R., Nonlinear chemical dynamics. Dalton Trans. 2003, (7), 1201 -

    work page doi:10.1146/annur 2003

  3. [3]

    (3) Migliore, A.; Nitzan, A., Nonlinear charge tran sport in redox molecular junctions: A Marcus perspective

    https://doi.org/10.1039/B210932H . (3) Migliore, A.; Nitzan, A., Nonlinear charge tran sport in redox molecular junctions: A Marcus perspective. ACS Nano 2011, 5 (8), 6669 -

    work page doi:10.1039/b210932h 2011

  4. [4]

    (4) Migliore, A.; Nitzan, A., Irreversibility and hysteresis in redox molecular conduction junctions

    https://doi.org/10.1021/nn202206e . (4) Migliore, A.; Nitzan, A., Irreversibility and hysteresis in redox molecular conduction junctions. J. Am. Chem. Soc. 2013, 135 (25), 9420 -

    work page doi:10.1021/nn202206e 2013

  5. [5]

    https://doi.org/10.1021/ja401336u . (5) Wang, Z.; Li, Z.; Li, C.; Ji, X.; Song, X.; Yu, X.; Wang, L.; Hu, W., Generic dynamic molecula r devices by quantitative non - steady - state proton/water - coupled electron transport kinetics. Proc. Natl. Acad. Sci. 2023, 120 (24), e2304506120. https://doi.org/10.1073/pnas.2304506120 . (6) Han, Y.; Nickle, C.; Zhan...

    work page doi:10.1021/ja401336u 2023

  6. [6]

    (7) Wang, Y.; Zhang, Q.; Astier, H

    https://doi.org/10.103 8/s41563 - 020 - 0697 - 5 . (7) Wang, Y.; Zhang, Q.; Astier, H. P.; Nickle, C.; Soni, S.; Alami, F. A.; Borrini, A.; Zhang, Z.; Honnigfort, C.; Braunschweig, B.; Leoncini, A.; Qu, D.; Han, Y.; Del Barco, E.; thompson, D.; nijhuis, C. A., Dynamic mol ecular switches with hysteretic negative differential conductance emulating synaptic...

    work page 2022

  7. [8]

    R.; Schwartz, J

    (12) Kandel, E. R.; Schwartz, J. H.; Jessell, T. M.; Siegelbaum, S.; Hudspeth, A. J.; Mack, S., Principles of neural science, 5th ed. McGraw - hill New York, 2000; Vol

    work page 2000

  8. [9]

    I.; Varona, P.; Selverston, A

    (14) Rabinovich, M. I.; Varona, P.; Selverston, A. I.; Abarbanel, H. D., Dynamical principles in neuroscience. Rev. Mod. Phys. 2006, 78 (4), 1213 –

    work page 2006

  9. [10]

    (15) Gerstner, W.; Kistler, W

    https://doi.org/10.1103/RevModPhys.78.1213 . (15) Gerstner, W.; Kistler, W. M.; Naud, R.; Paninski, L., Neuronal dynamics: From single neurons to networks and models of cognition . Cambridge University Press: Cambridge,

    work page doi:10.1103/revmodphys.78.1213

  10. [11]

    Neuron 2017, 95 (2), 245 -

    (16) Hassabis, D.; Kumaran, D.; Summerfield, C.; Botvinick, M., Neuroscience - inspired artificial intelligence. Neuron 2017, 95 (2), 245 -

    work page 2017

  11. [12]

    (17) Hopfield, J

    https://doi.org/10.1016/j.neuron.2017.06.011 . (17) Hopfield, J. J., Neural networks and physical systems with emergent collective computational abilities. Proc. Natl. Acad. Sci. 1982, 79 (8), 2554 -

    work page doi:10.1016/j.neuron.2017.06.011 2017

  12. [13]

    Christopher G

    https://doi.org/10.1073/PNAS.79.8.2554 . (18) Mark ović, D.; Mizrahi, A.; Querlioz, D.; Grollier, J. J., Physics for neuromorphic computing. Nat. Rev. Phys. 2020, 2 , 499 -

    work page doi:10.1073/pnas.79.8.2554 2020

  13. [14]

    (19) Rabinovich, M.; Huerta, R.; Laurent, G., Transient dynamics for neural processing

    https://doi.org/10.1038/s42254 - 020 - 0208 - 2 . (19) Rabinovich, M.; Huerta, R.; Laurent, G., Transient dynamics for neural processing. Science Advances 2008, 321 (5885), 48 -

    work page doi:10.1038/s42254 2008

  14. [15]

    (20) Wang, Z.; Joshi, S.; Savel’ev, S

    https://doi.org/10.1126/science.1155564 . (20) Wang, Z.; Joshi, S.; Savel’ev, S. E.; Jiang, H.; Midya, R.; Lin, P.; Hu, M.; Ge, N.; Strachan, J. P.; Li, Z.; Wu, Q.; Barnell, M.; Li, G.; Xin, H.; Williams, R. S.; Xia, Q.; Yang, J. J., Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing. Nat. Mater. 2017, 16 , 101 -

    work page doi:10.1126/science.1155564 2017

  15. [16]

    (21) Du, C.; Cai, F.; Zidan, M

    https://doi.org/10.1038/nmat4756 . (21) Du, C.; Cai, F.; Zidan, M. A.; Ma, W.; Lee, S. H.; Lu, W. D., Reservoir computing using dynamic memristors for temporal information processing. Nat. Commu n. 2017, 8 ,

    work page doi:10.1038/nmat4756 2017

  16. [18]

    (23) Usami, Y.; van de Ve n, B.; Mathew, D

    https://doi.org/10.1038/s41563 - 021 - 01099 - 9 . (23) Usami, Y.; van de Ve n, B.; Mathew, D. G.; Chen, T.; Kotooka, T.; Kawashima, Y.; Tanaka, Y.; Otsuka, Y.; Ohoyama, H.; Tamukoh, H.; Tanaka, G.; van der wiel, W. G.; Mastsumoto, T., In ‐ materio reservoir computing in a sul fonated polyaniline network. Adv. Mater. 2021, 3 3 (48), 2102688. https://doi.o...

    work page doi:10.1038/s41563 2021

  17. [19]

    Science 2004, 304 (5667), 78 -

    (26) Jaeger, H.; Haas, H., Harnessing nonlinearity: Predicting chaotic systems and saving energy in wireless communication. Science 2004, 304 (5667), 78 -

    work page 2004

  18. [20]

    (27) Tanaka, G.; Yamane, T

    https://doi.org/10.1126/science.1091277 . (27) Tanaka, G.; Yamane, T. ; Héroux, J. B.; Nakane, R.; Kanazawa, N.; Takeda, S.; Numata, H.; Nakano, D.; Hirose, A., Recent advances in physical reservoir computing: A review. Neural Netw. 2019, 115 , 100 -

    work page doi:10.1126/science.1091277 2019

  19. [21]

    (28) Zhong, Y.; Tang, J.; Li, X.; Gao, B.; Qian, H.; Wu, H., Dynamic memristor - based reservoir computing for high - efficiency temporal signal processing

    https: //doi.org/10.1016/j.neunet.2019.03.005 . (28) Zhong, Y.; Tang, J.; Li, X.; Gao, B.; Qian, H.; Wu, H., Dynamic memristor - based reservoir computing for high - efficiency temporal signal processing. Nat. Commun. 2021, 12 ,

    work page doi:10.1016/j.neunet.2019.03.005 2019

  20. [26]

    https://doi.or g/10.1038/ncomms1476 . (35) Cucchi, M.; Parker, D.; Stavrinidou, E.; Gkoupidenis, P.; Kleemann, H., In Liquido Computation with Electrochemical Transistors and Mixed Conductors for Intelligent Bioelectronics. Adv. Mater. 2023, 35 (15), 2209516. https://doi.org/10.1002/adma.202209516 . (36) Pecqueur, S.; Mastropasqua Talamo, M.; Guérin, D.; ...

    work page doi:10.1038/ncomms1476 2023

  21. [27]

    (39) Shougat, M

    https://doi.org/10.1088/0957 - 4484/24/38/384004 . (39) Shougat, M. R. E. U.; Li, X.; Shao, S.; McGarvey, K.; Perkins, E., Hopf physical reservoir computer for reconfigurable sound recognition. Sci. Rep. 2023, 13 ,

    work page doi:10.1088/0957 2023

  22. [28]

    https://doi.org/10.1038/s41598 - 023 - 35760 - x . (40) Liu, K.; Zhang, T.; Dang, B.; Bao, L.; Xu, L.; Cheng, C.; Yang, Z.; Huang, R.; Yang, Y., An optoelectronic synapse based on α - In2Se3 with controllable temporal dynamics for multimode and multiscale reservoir computing. Nat. Electr on. 2022, 5 , 761 -

    work page doi:10.1038/s41598 2022

  23. [29]

    (41) Tan, H.; van Dijken, S., Dynamic machine vision with retinomorphic photomemristor - reservoir computing

    https://doi.org/10.1038/s41928 - 022 - 00847 - 2 . (41) Tan, H.; van Dijken, S., Dynamic machine vision with retinomorphic photomemristor - reservoir computing. Nat. Commun. 2023, 14 ,

    work page doi:10.1038/s41928 2023

  24. [32]

    (4 6 ) Riou, M.; Araujo, F

    https://doi.org/10.1038/s41467 - 023 - 3 5817 - 5 . (4 6 ) Riou, M.; Araujo, F. A.; Torrejon, J.; Tsunegi, S.; Khalsa, G.; Querlioz, D.; Bortolotti, P.; Cros, V.; Yakushiji, K.; Fukushima, A.; Kubota, H.; Yuasa, S.; Grollier, J.; Stiles, M. D. In Neuromorphic computing through time - multipl exing with a spin - torque nano - oscillator , 2017 IEEE Interna...

    work page doi:10.1038/s41467 2017

  25. [33]

    1109/IEDM.2017.8268505

    https://doi.org/10. 1109/IEDM.2017.8268505 . (4 7 ) Paquot, Y.; Duport, F.; Smerieri, A.; Dambre, J.; Schrauwen, B.; Haelterman, M.; Massar, S., Optoelectronic reservoir computing. Sci. Rep. 2012, 2 ,

    work page arXiv 2017

  26. [34]

    1038/srep00287

    https://doi.org/10. 1038/srep00287 . (4 8 ) Verstraeten, D.; Schrauwen, B.; Stroobandt, D. In Reservoir - based techniques for speech recognition , The 2006 IEEE International Joint Conference on Neural Network Proceedings., Vancouver, BC, Canada, IEEE: Vancouver, BC, Canada, 2006 ; pp 1050 -

    work page 2006

  27. [35]

    (4 9 ) Lukoševičius, M.; Jaeger, H., Reservoir computing approaches to recurrent neural network training

    https://doi.org/10.1109/IJCNN.2006.246804 . (4 9 ) Lukoševičius, M.; Jaeger, H., Reservoir computing approaches to recurrent neural network training. Comput. Sci. Rev. 2009, 3 (3), 127 -

    work page doi:10.1109/ijcnn.2006.246804 2006

  28. [36]

    ( 50 ) Appeltant, L.; Van der Sande, G.; Danckaert, J.; Fischer, I., Constructing optimized binary masks for reservoir computing with delay systems

    https://doi.org/10.1016/j.cosrev.2009.03.005 . ( 50 ) Appeltant, L.; Van der Sande, G.; Danckaert, J.; Fischer, I., Constructing optimized binary masks for reservoir computing with delay systems. Sci. Rep. 2014, 4 ,

    work page doi:10.1016/j.cosrev.2009.03.005 2009

  29. [37]

    (5 1 ) Mackey, M

    https://doi.org/10.1038/srep03629 . (5 1 ) Mackey, M. C.; Glass, L., Oscillation and chaos in physiological control systems. Science 1977, 197 (4300), 287 -

    work page doi:10.1038/srep03629 1977

  30. [38]

    (5 2 ) Hunt, B

    https://doi.org/10.1126/science.267326 . (5 2 ) Hunt, B. R.; Kennedy, J. A.; Li, T. - Y.; Nusse, H. E., The theory of chaotic attractors . Springer Science & Business Media:

    work page doi:10.1126/science.267326

  31. [39]

    N., Deterministic nonperiodic flow

    (5 3 ) Lorenz, E. N., Deterministic nonperiodic flow. J. Atmos. Sci. 1963, 20 (2), 130 -

    work page 1963

  32. [40]

    (5 4 ) Han, X.; Zhao, Y.; Small, M., Revisiting the memory capacity in reservoir computing of directed acyclic network

    https://doi.org/10.1175/1520 - 0469(1963)020 <0130:DNF>2.0.CO;2. (5 4 ) Han, X.; Zhao, Y.; Small, M., Revisiting the memory capacity in reservoir computing of directed acyclic network. Chaos 2021, 31 (3), 033106. https://doi.org/10.1063/5.0040251 . (5 5 ) Jaeger, H.; Doorakkers, D.; Lawrence, C.; Indiveri, G., Dimensions of timescales in neuromorphic comp...

    work page doi:10.1175/1520 1963

  33. [41]

    F.; Parlos, A

    (5 7 ) Atiya, A. F.; Parlos, A. G., New results on recurrent network training: unifying the algorithms and accelerating convergence. IEEE Trans. Neural Netw. 2000, 11 (3), 697 -

    work page 2000

  34. [42]

    (5 8 ) Fukuda, K.; Horio, Y., Analysis of dynamics in chaotic neural network reservoirs: Time - series prediction tasks

    https://doi.org/10.1109/72.846741 . (5 8 ) Fukuda, K.; Horio, Y., Analysis of dynamics in chaotic neural network reservoirs: Time - series prediction tasks. Nonlinear Theory Appl. IEICE 2021, 12 (4), 639 -

    work page doi:10.1109/72.846741 2021

  35. [43]

    https://doi.org/10.1587/nolta.12.639 . (5 9 ) Kan, S.; Nakajima, K.; Takeshima, Y.; Asai, T.; Kuwahara, Y.; Akai - Kasaya, M., Simple reservoir computing capitalizing on the nonlinear response of materials: theory and physical implementatio ns. Phys. Rev. Appl. 2021, 15 (2), 024030. https://doi.org/10.1103/physrevapplied.15.024030

    work page doi:10.1587/nolta.12.639 2021