Pith. sign in

REVIEW 3 major objections 4 minor 66 references

Stochastic resonance in a metal-oxide memristive device

T0 review · 3 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Optimal noise peaks a memristor's response and steadies its switching.

desk verdict Plausible but under-specified experimental claim of stochastic resonance in a memristor; the SNR curve needs a clear protocol and error bars before the peak can be trusted. read the letter →

arxiv 2412.05307 v1 pith:ZMG4BGRK submitted 2024-11-27 cond-mat.mes-hall

A.N. Mikhaylov , D.V. Guseinov , A.I. Belov , D.S. Korolev , V.A. Shishmakova , M.N. Koryazhkina , D.O. Filatov , O.N. Gorshkov
show 8 more authors
D. Maldonado F.J. Alonso J.B. Roldan A.V. Krichigin N.V. Agudov A.A. Dubkov A. Carollo B. Spagnolo
This is my paper · ORCID
classification cond-mat.mes-hall
keywords memristorstochasticresonanceresistiveswitchingyttria-stabilizedzirconiatantalumoxidesignal-to-noiserationoise-inducedcoarse-grainedmodel
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to show that random noise, usually a nuisance in a metal-oxide memristor (a two-terminal resistor whose resistance remembers the voltage it has seen), can be put to work. The authors drive a $\mathrm{ZrO_2(Y)/Ta_2O_5}$ memristor with a 10 Hz sine wave too weak to switch it, add white Gaussian voltage noise of increasing intensity, and find that the device's signal-to-noise ratio (SNR) at the driving frequency first falls, then rises to a clear maximum near $\theta_\zeta \approx 10$–$12$ $\mathrm{mV}^2\mathrm{s}$, and then falls again — the fingerprint of stochastic resonance. At that optimal noise level the resistive-switching hysteresis loop is largest, the ratio of high- to low-resistance state currents is maximized, and cycle-to-cycle variability of the resistive states is sharply reduced. If correct, this means noise is not merely a stability problem for memristive devices but a tunable resource: a deliberately chosen noise level can stabilize switching under a sub-threshold drive.

What carries the argument

The argument runs on a coarse-grained stochastic memristor model in which oxygen vacancies hop between trapping sites under a periodic potential tilted by the applied voltage. The hopping is described by a Langevin equation for a diffusing particle; the external white Gaussian voltage noise $\zeta(t)$ is added to the deterministic drive $V_0(t)$, and because both noise sources enter the same linear way, they merge into one effective noise intensity $\theta_\nu = (q/\varepsilon L)^2 \theta_\zeta + \theta_\xi$. That folding turns the external noise into an apparent temperature increase, which lowers the Kramers switching time $\tau = \tau_0 \exp(E_a/\theta_\nu)$ and makes the threshold-like resistance profile switchable. The model's Fokker–Planck equation for the defect concentration, together with a thresholded resistivity function and an effective drift coefficient $\sinh(BV_0/2\theta_\nu)$, produces simulated I–V loops that track the experimental trend of growing-then-shrinking hysteresis.

What would settle it

Measure the SNR-versus-noise curve with band-limited noise of the same variance but different spectral content, or at several driving frequencies, and compare with the model's effective-temperature prediction. If the peak position moves when only the noise spectrum changes, or if the SNR peak disappears when the drive frequency is changed while the model says it should persist, the equivalence between external noise and thermal noise is falsified; a quantitative fit of the simulated I–V loops to the measured ones would also settle whether the mechanism is the only one at work.

Watch

Extended reading notes

Core claim

The central claim is that a metal-oxide memristive device exhibits classic stochastic resonance when white Gaussian noise is superimposed on a sub-threshold sinusoidal driving voltage. Measured as the signal-to-noise ratio of the memristance at the 10 Hz driving harmonic, the response is non-monotonic in the external noise intensity: the SNR decreases for very weak noise, then grows to a maximum near $\theta_\zeta \approx 10$–$12$ $\mathrm{mV}^2\mathrm{s}$, and then decreases again at higher intensities. In the same range, the area of the I–V hysteresis loop first increases and later shrinks, the completeness of SET/RESET switching improves, and the variation of the resistive states after switching drops and saturates. The authors interpret all of this with a stochastic memristor model in which the external voltage noise adds to the thermal noise intensity, so that increasing external noise is equivalent to raising the effective temperature; the observed regularities are presented as evidence that noise plays a constructive role in nonequilibrium memristive systems.

Load-bearing premise

The whole mechanistic story rests on treating added external voltage noise as equivalent to raising the device temperature, since the model only sees the combined noise intensity; if the real device reacts to fast noise spikes through Joule-heating inertia or ion-migration delays that the model ignores, the claimed cause of the SNR peak would not follow from the data.

Editorial extensions

If this is right

  • At the optimal noise intensity the device can be switched with a drive amplitude below the usual threshold, so noise can lower the voltage budget for resistive switching.
  • The signal-to-noise ratio peak gives a measurable operating point: adding noise around $\theta_\zeta \approx 10$–$12$ $\mathrm{mV}^2\mathrm{s}$ maximizes the memristance response at the driving frequency.
  • Cycle-to-cycle variability of the resistive states shrinks with increasing noise up to saturation, implying noise can be used to stabilize device statistics rather than only disturb them.
  • The non-monotonic hysteresis-loop area provides an experimentally accessible signature of stochastic resonance that does not require spectral analysis.
  • Time-series models fitted to the reset voltage at different noise intensities allow forecasting of switching-voltage series and quantify how noise changes the memory of past cycles.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the effective-temperature equivalence holds quantitatively, the optimal noise intensity should shift with the driving frequency and with ambient temperature; measuring that shift would be a direct test of the mechanism.
  • The same strategy may transfer to other filamentary metal-oxide memristors, but the optimal noise level likely scales with the material's activation energy, so device-specific tuning would be required.
  • Used deliberately, noise could serve as a free control parameter in neuromorphic circuits, where tunable stochasticity in switching is often desirable; the paper does not explore this application.
  • Because high-frequency noise spikes are assumed not to modify the filament directly, the mechanism implies a bandwidth limit: noise with spectral content faster than the thermal and ionic response times should stop behaving like an effective temperature.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper reports an experimental and theoretical study of stochastic resonance (SR) in a ZrO2(Y)/Ta2O5-based metal-oxide memristive device. The authors apply a sub-threshold 10 Hz sinusoidal voltage with superimposed white Gaussian noise and characterize the response through I-V hysteresis loops, time-series statistics of switching parameters, and the spectral signal-to-noise ratio (SNR) of the memristance at the driving frequency. They report that the I-V hysteresis loop area first grows and then shrinks with noise intensity, the ratio of high- to low-resistance-state currents is non-monotonic with a maximum near θζ = 14.43 mV²s, resistive-state variability decreases, and the SNR at 10 Hz has a maximum near θζ ≈ 10–12 mV²s. These observations are interpreted with a coarse-grained stochastic memristor model in which external voltage noise is folded into an effective thermal-noise intensity, and the switching time relative to the driving period is used to explain the non-monotonic hysteresis and SNR behavior.

Significance. If the main claims hold, the paper provides a useful experimental demonstration that external noise can stabilize and regularize resistive switching in a realistic metal-oxide memristor, going beyond earlier theoretical and preliminary experimental studies. The paper's strengths include the use of multiple complementary observables (I-V loop area, resistance ratio, state variability, SNR, autocorrelation functions) and the connection to a stochastic model developed and calibrated in previous work. However, the quantitative evidence for the headline SR claim rests on a single unreplicated SNR curve whose extraction procedure is not documented, and the model-based interpretation relies on an effective-temperature equivalence that is asserted rather than validated for the specific device and frequency range. These gaps currently prevent the paper from fully establishing the SR mechanism.

major comments (3)
  1. [§4.3, Fig. 12] The SNR curve in Fig. 12 is the central quantitative evidence for the SR claim, but the manuscript does not specify how the memristance Rm(t) is extracted from the measured current through the 100 Ω series resistor, nor does it give the SNR formula, the FFT windowing or frequency resolution, the number of averaged cycles, or the noise-floor subtraction procedure. The phrase "relative to the level of noise" is too vague to reproduce the measurement. In addition, Fig. 12 has no error bars and appears to be from a single device and a single measurement series. Without this information, the non-monotonic peak near θζ ≈ 10–12 mV²s could be produced or destroyed by spectral leakage from the switching transients, by division by near-zero spectral components, or by noise-floor contamination. Please provide a complete extraction protocol and include error bars or replicate measurements.
  2. [§3, Eqs. (3.3)–(3.5)] The model's central assumption is that external white Gaussian voltage noise can be absorbed into the effective thermal noise intensity via Eq. (3.5), so that increasing the external noise is equivalent to raising the device temperature. This equivalence requires that high-frequency voltage fluctuations affect defect hopping in exactly the same way as thermal fluctuations, which is not obvious for a device with Joule-heating inertia and ion-migration delays. Indeed, §4.2 states that for noise intensities below 34.16 mV²s the voltage spikes "do not alter much the CFs since the thermal inertia and the ion migration delays connected to the RS processes avoid a great modification in the CF nature." That statement is in tension with the effective-temperature picture used to explain the observed SR-like behavior. The authors should justify the frequency and amplitude range over which Eq. (3.5) applies to this device, or explicitly delimit the model's interpretative role.
  3. [§4.1, Fig. 5] The comparison between measured and simulated I-V characteristics in Fig. 5 is only qualitative. The model parameters are taken from the prior calibration in Ref. [34], and no quantitative metric (e.g., hysteresis-loop area versus θζ, or switching-voltage distributions) is compared between model and experiment. Moreover, the simulated curves are ensemble-averaged while the experimental curves are single-realization measurements, so the two are not directly comparable in a statistical sense. Since the model is used to explain the observed regularities through the switching-time argument in §4.1, the paper should either provide a quantitative model-experiment comparison or clearly present the model as an illustrative rather than validated mechanism.
minor comments (4)
  1. [§2] The paragraph describing the classical SR experiment and the measurement setup appears twice verbatim in Section 2; one copy should be removed.
  2. [§3] The equations in Section 3 are heavily garbled in the manuscript text (for example, Eq. (3.1) and the definitions of symbols are unreadable), which makes it difficult to verify the derivation of Eqs. (3.3)–(3.5). The final version should contain correctly rendered mathematics.
  3. [Title and Abstract] The title contains a spacing typo ("de vice") and the abstract text says "Qme series" instead of "time series"; these should be corrected.
  4. [§2] The text states that the sampling rate of 6.553 kHz corresponds to a correlation time of 1.526×10⁻⁴ s; this is the sampling interval, not a correlation time, and the wording should be fixed.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation: the claimed SNR peak is an experimental measurement independent of the fitted stochastic model, so the central result does not reduce to its inputs.

full rationale

The central claim is the experimentally obtained non-monotonic SNR curve in Fig. 12. The text states: 'If we consider the SNR measured for the main harmonic at 10 Hz in the spectrum of Rm(t) relative to the level of noise, we obtain the dependence of SNR on external intensity shown in Fig. 12.' This quantity is measured from the device response, not computed from the model. Equation (3.5), which folds external voltage noise into an equivalent thermal-noise intensity, is an interpretive modeling step used for the simulated I-V curves; no model parameter is fitted to the SNR curve and the SNR peak is not defined into existence by Eqs. (3.1)-(3.14). The stochastic model and TSSA tools are taken from the authors' own prior work (Refs. [34], [53], [54]), but these self-citations are used for qualitative interpretation ('The variation of RS properties can be explained, when we take into account the changing difference between the two time scales'), not as the source of the measured peak. The absence of an explicit Rm(t) extraction protocol and error bars in Fig. 12 is a reproducibility concern, not circularity: even if that protocol is incomplete, the measured claim would be weak evidence, but it would not be equivalent to the paper's inputs. No specific reduction of a prediction to a fit or to a self-citation chain can be quoted.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claim does not introduce new physical entities; it relies on a stochastic model with several free parameters (D0, beta, B) inherited from prior calibration, on the assumption that external noise equals effective thermal noise, and on the boundary conditions and threshold resistivity chosen by the authors. The experimental SNR peak is independent of these model choices, but the mechanistic interpretation depends on them.

free parameters (4)
  • D0 (diffusion coefficient) = 6e-13 cm^2/s
    Used in Eqs. (3.8)-(3.10) for effective drift and diffusion; value is taken from prior calibration in Ref. [34], not fitted to the data in this paper.
  • beta (dimensionless barrier height Ea/theta_nu) = 25 (for theta_nu = theta_xi)
    Sets the Kramers time scale in Eq. (3.10) and is used in all model simulations in Fig. 5; adopted from the calibrated model of Ref. [34].
  • B (fitting coefficient, B = q*a/(epsilon L)) = not stated
    Appears in alpha = B V0 / (2 theta_nu) in Fig. 2 and in the effective drift coefficient; the text calls B a fitting coefficient but does not give its value or fitting procedure.
  • n_th (threshold defect concentration for resistivity step) = N/2
    Defines the abrupt resistivity change in Eq. (3.13); chosen by assumption rather than measured.
assumptions (4)
  • domain assumption The coarse-grained stochastic memristor model of Ref. [34] (Fokker-Planck equation for defect concentration with Kramers escape rates) accurately describes the device dynamics.
    Invoked in Section 3 and used to simulate all I-V curves in Fig. 5; this paper does not re-derive or validate the model, it imports it.
  • domain assumption External voltage noise adds a white Gaussian term that is equivalent to an increase in thermal noise intensity, so theta_nu = (q^2/(epsilon^2 L^2)) theta_zeta + theta_xi.
    Key step in Eqs. (3.3)-(3.5); requires that the potential U(x,V) is linear in V and that the noise is ideal white Gaussian. This equivalence underlies the entire interpretation of the experiments.
  • domain assumption The boundary condition for the defect concentration is fixed (N=100%) at the TiN/Ti electrode and reflecting at the Au electrode.
    Used to close the FPE in Eqs. (3.11)-(3.12); no independent evidence is given for this choice.
  • ad hoc to paper The resistivity of the device is a threshold-like function of defect concentration, with the threshold at n_th = N/2.
    Given in Eq. (3.13); this step function is a modeling choice that provides the nonlinear threshold behavior needed for stochastic resonance, but its quantitative form is not derived from material properties.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Stochastic resonance in a metal-oxide memristive device." pith.science (2026). https://pith.science/paper/ZMG4BGRK

@misc{pith2026241205307,
  author       = {Pith},
  title        = {Pith review of: Stochastic resonance in a metal-oxide memristive device},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZMG4BGRK}},
  note         = {Machine review of arXiv:2412.05307}
}
read the original abstract

The stochastic resonance phenomenon has been studied experimentally and theoretically for a state-of-art metal-oxide memristive device based on yttria-stabilized zirconium dioxide and tantalum pentoxide, which exhibits bipolar filamentary resistive switching of anionic type. The effect of white Gaussian noise superimposed on the sub-threshold sinusoidal driving signal is analyzed through the time series statistics of the resistive switching parameters, the spectral response to a periodic perturbation and the signal-to-noise ratio at the output of the nonlinear system. The stabilized resistive switching and the increased memristance response are revealed in the observed regularities at an optimal noise intensity corresponding to the stochastic resonance phenomenon and interpreted using a stochastic memristor model taking into account an external noise source added to the control voltage. The obtained results clearly show that noise and fluctuations can play a constructive role in nonlinear memristive systems far from equilibrium.

Figures

Figures reproduced from arXiv: 2412.05307 by the authors.

Figure 1
Figure 1. Input voltage vs. measurement time for a single cycle applied to the mem- ristive device under study for two noise intensities. To conduct such experiment, an external Gaussian noise signal was generated using the ADSViewer-2 (v.015) pseudorandom number generator [59] and characterized by the voltage amplitude (maximum variation in voltage values). This noise signal was superimposed on a sinusoidal voltage signal, t… view at source ↗
Figure 2
Figure 2. Dimensionless potential profile (3.2) for V0= 0 and V0 > 0, where β = Ea/θν is the dimensionless height of potential barrier under V0 = 0 and α = BV0/2θν is the value of barrier height variation under V0 ≠ 0 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Diagram showing the smoothed coarse-grained concentration of defects n(x, t ) with the equivalent electric circuit of the memristor model. The boundary conditions for FPE (3.6) should be specified according to the material of electrodes. In our case we consider relative concentration of defects N=100% at the easily oxidizable TiN/Ti electrode #0,  A , (3.11) and reflecting boundary conditions for the inert Au ele… view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: I-V characteristics of the memristive device under study measured before each application of a periodic sinusoidal signal with the amplitude of 1 V and dif- ferent noise intensities. 4. Results and Discussion 4.1. Resistive Switching Regularities In [PITH_FULL_IMAGE:f…
Figure 5
Figure 5. Figure 5: I-V characteristics of the memristive device under study measured (left panels) and simulated (right panels) under the amplitude of driving signal 1 V and frequency 10 Hz for the growing intensity of external noise [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Time series of the current response of memristive device under study on different time scales, measured for different noise intensities added to a sinusoidal driving signal with amplitude of 1 V [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: The LRS and HRS current ratio (ILRS / IHRS ), as well as the variation of current after switching to HRS (∆IHRS ) vs. the noise intensity [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: Experimental current and voltage versus time elapsed (0.1s/cycle) for three consecutive RS cycles in a long series for the memristive device under study for a noise intensity of 0.09 mV2 s. According to the coarse-grained stochastic model [34], the switching time (or r…
Figure 9
Figure 9. Figure 9: Experimental RESET voltages obtained for each measured cycle for the RS series analyzed at different noise intensities. 4.2. Time Series Statistical Analysis The TSSA has been employed to characterize the statistical features of the memristive device operation variable…
Figure 10
Figure 10. Figure 10: ACF (left panels) and PACF (right panels) vs. cycle lag (distance apart in cycles within a RS series; for a cycle lag 1, the ACF and PACF of consecutive cycles are measured and so on) for the Vreset series described in [PITH_FULL_IMAGE:figures/full_fig_p010_10.png]
Figure 11
Figure 11. Figure 11: Typical power spectrum of the memristance Rm (t) of device under study measured for a noise intensity of 8.56 mV2 s. 4.3. Spectral Response and SNR The most informative parameter of RS is the value of memristor resistance (memristance). As it was described above in th…
Figure 12
Figure 12. Figure 12: Memristance SNR vs. noise intensity experimentally obtained for the memristive device under study. 5. Conclusions In the present work, we have investigated the main features of RS parameters variation under the application of a Gaussian noise signal added to the drivi…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

66 extracted references · 49 canonical work pages

  1. [34]

    Nonstationary distri butions and relaxation times in a stochastic model of memristor

    Agudov NV, Safonov AV, Krichigin AV, Kharcheva AA, Dubkov AA, Valenti D, Guseinov DV, Belov AI, Mikhay lov AN, Carollo A, Spagnolo B. Nonstationary distri butions and relaxation times in a stochastic model of memristor . J Stat Mech Theory Exp 2020;2020:024003. doi: 10.1088/1742-5468/ab684a

    work page doi:10.1088/1742-5468/ab684a 2020

  2. [1]

    Resistive Switching: From Funda mentals of Nanoionic Re- dox Processes to Memristiv e Device Applications

    Ielmini D, Waser R. Resistive Switching: From Funda mentals of Nanoionic Re- dox Processes to Memristiv e Device Applications. Weinheim, Germany: Wi- ley-V CH Verlag GmbH & Co KGaA; 2016

    work page 2016

  3. [2]

    Nanoscale resistive switching devices for mem- ory and computing applications

    Lee SH, Zhu X, Lu WD. Nanoscale resistive switching devices for mem- ory and computing applications. N ano Res 2020;13:1228–43. doi: 10.1007/ s12274- 020- 2616- 0

    work page 2020

  4. [3]

    Resistive switching materials for informatio n processing

    Wang Z, Wu H, Burr GW, Hwang CS, Wang KL, Xia Q, Ya ng JJ. Resistive switching materials for informatio n processing. Nat Rev Mater 2020;5:173–95. doi: 10.1038/s41578-019-0159- 3

    work page doi:10.1038/s41578-019-0159- 2020

  5. [4]

    Brain-inspired comput- ing with memristors: Chal lenges in devices, circuits, and systems

    Zhang Y, Wang Z, Zhu J, Yang Y, Rao M, Song W, et a l. Brain-inspired comput- ing with memristors: Chal lenges in devices, circuits, and systems. Appl Phys Rev 2020;7:011308. doi: 10.1063/1.5124027

    work page doi:10.1063/1.5124027 2020

  6. [5]

    Memristive Devices for Neuromorphic Appl ications: Comparative Analysis

    Erokhin V. Memristive Devices for Neuromorphic Appl ications: Comparative Analysis. Bionanoscience 2020 ;10:834–47. doi: 10.1007/s12668- 020-00795- 1

    work page doi:10.1007/s12668- 2020

  7. [6]

    Neurohybrid memristive CMOS- inte grated systems for biosensors and neuroprosthetics

    Mikhaylov A, Pimashkin A, Pigareva Y, Gerasimova S, Gryaznov E, Shchanikov S, Zuev A, Talanov M, Lavro v I, Demin V, Erokhin V, Lobov S, Mukhina I, Kazant sev V, Wu H, Spagnolo B. Neurohybrid memristive CMOS- inte grated systems for biosensors and neuroprosthetics. Front Neurosci 2020 14:358. doi: 10.3389/fnins.2020.00358

    arXiv 2020

  8. [7]

    Designing cr ystallization in phase- change materials for univer sal memory and neuro-inspired computing

    Zhang W, Mazzarello R, Wuttig M, Ma E. Designing cr ystallization in phase- change materials for univer sal memory and neuro-inspired computing. Nat Rev Ma ter 2019;4:150–68. doi: 10.1038/s41578-018-0076- x

    work page doi:10.1038/s41578-018-0076- 2019

Show all 66 references
  1. [8]

    Training and operation of a n integrated neuromorphic network based on metal-ox ide memristors

    Prezioso M, Merrikh-Bayat F, Hoskins BD, Adam GC, L ikharev KK, Strukov DB. Training and operation of a n integrated neuromorphic network based on metal-ox ide memristors. Nature 2015;521:61–4. doi: 10.1038/nature14441

    2015 doi

  2. [9]

    Neuromorphic computing using non-vo latile memory

    Burr GW, et al. Neuromorphic computing using non-vo latile memory. Adv. Phys. X 2016;2:89–124. doi: 10.1080/23746149.2016.1259585

    2016

  3. [10]

    In-memory computing with resis tive switching devices

    Ielmini D, Wong HSP. In-memory computing with resis tive switching devices. Nat Electron 2018;1:333–43. doi: 10.1038/s41928-018-0092- 2

    2018 doi

  4. [11]

    Neuro-inspired computing with emerging nonvol atile memorys

    Yu S. Neuro-inspired computing with emerging nonvol atile memorys. Proc. IEEE 2018;106:260–85. doi: 10.1109/JPROC.2018.2790840

    2018

  5. [12]

    The future of electro nics based on memristive systems

    Zidan MA, Strachan JP, Lu WD. The future of electro nics based on memristive systems. Nat Electron 2018 ;1:22–9. doi: 10.1038/s41928-017- 0006-8

    2018 doi

  6. [13]

    Equivalent-accuracy accelerated neural-network training us- ing analogue memory

    Ambrogio S, et al. Equivalent-accuracy accelerated neural-network training us- ing analogue memory. Na ture 2018;558:60–7. doi: 10.1038/s41586-018-0180- 5

    2018 doi

  7. [14]

    Nanoionics-based resistive switchi ng memories

    Waser R, Aono M. Nanoionics-based resistive switchi ng memories. Nat Mater 2007;6:833–40. doi: 10.1038/nmat2023

    2007 doi

  8. [15]

    Recommended Method s to Study Resistive Switching Devices

    Lanza M, Wong H-SP, Pop E, Ielmini D, Strukov D, Re gan BC, Larcher L, Vil- lena MA. Recommended Method s to Study Resistive Switching Devices. Adv Electro n Mater 2019;5:1800143. doi: 10.1002/aelm.201800143

    2019 doi

  9. [16]

    The missing memristor found

    Strukov DB, Snider GS, Stewart DR, Williams RS. The missing memristor found. Nature 2008;453:80–3. doi : 10.1038/nature06932

    2008 doi

  10. [17]

    Memristor-The missing circuit element

    Chua L. Memristor-The missing circuit element. IEEE Trans Circuit Theory 1971;18:507–19. doi: 10.1109/TCT.1971.1083337

    1971

  11. [18]

    Memristive devices and syste ms

    Chua LO, Kang Sung Mo. Memristive devices and syste ms. Proc IEEE 1976;64:209–23. doi: 10.1109/PROC.1976.10092

    1976

  12. [19]

    Resistive switching memories based on me tal oxides: Mecha- nisms, reliability and scaling

    Ielmini D. Resistive switching memories based on me tal oxides: Mecha- nisms, reliability and scaling. Semicond Sci Technol 2016;31:1–25. doi: 10.1088/ 0268- 1242/31/6/063002

    2016

  13. [20]

    Optimization of impurity profile for p-n junc- tion in heterostructures

    Pankratov E L, Spagnolo B. Optimization of impurity profile for p-n junc- tion in heterostructures. Eur. Phys. J. B 200 5;46:15–19. doi: 10.1140/epjb/ e2005-00233-1

    doi

  14. [21]

    Thermophoresi s/diffusion as a plausible mechanism for unipolar r esistive switching in metal–oxide–metal memristors Appl

    Strukov D B, Alibart F, Williams R S. Thermophoresi s/diffusion as a plausible mechanism for unipolar r esistive switching in metal–oxide–metal memristors Appl. Phys. A 2012;107:509–18. doi: 10.1007/s00339-012- 6902- x

    2012 doi

  15. [23]

    Reversible di electric breakdown in h-BN stacks: a statistical st udy of the switching voltages

    Roldan JB, Maldonado D, Jimenez-Molinos F, Acal C, Ruiz-Castro JE, Aguil- era AM, et al. Reversible di electric breakdown in h-BN stacks: a statistical st udy of the switching voltages. IEEE Int Reliab Phys Symp 2 020:1–5 2020- April. doi: 10.1109/IRPS45951.2020.9129147

    2020

  16. [24]

    Field- and irradiation-induce d phenomena in memristive nano- materials

    Mikhaylov AN, Gryaznov EG, Belov AI, Korolev DS, Sh arapov AN, Guseinov DV, Tetelbaum DI, Tikhov SV, Ma lekhonova NV, Bobrov AI, Pavlov DA, Gerasi- mova SA , Kazantsev VB, Agudov NV, Dubkov AA, Rosário CMM, So bolev NA, Spagnolo B. Field- and irradiation-induce d phenomena in ...

    2016 doi

  17. [25]

    An overview of materia ls issues in resistive ran- dom access memory

    Zhu L, Zhou J, Guo Z, Sun Z. An overview of materia ls issues in resistive ran- dom access memory. J Ma ter 2015;1:285–95. doi: 10.1016/j.jmat.2015.07.009

    2015 doi

  18. [26]

    Impact of ultra- thin Al 2 O3-y layers on TiO 2-x ReRAM switching characteristics

    Trapatseli M, Cortese S, Serb A, Khiat A, Prodromak is T. Impact of ultra- thin Al 2 O3-y layers on TiO 2-x ReRAM switching characteristics. J Appl Phys 2017;1 21:1–9. doi: 10.1063/1.4983006

    2017 doi

  19. [27]

    Mul- tilayer Metal Oxide Memr istive Device with Stabilized Resistive Switching

    Mikhaylov A, Belov A, Korolev D, Antonov I, Kotomin a V, Kotina A, et al. Mul- tilayer Metal Oxide Memr istive Device with Stabilized Resistive Switching. Adv Mater Technol 2020;5:1900607. doi: 10.1002/admt.201900607

    2020 doi

  20. [28]

    High prec ision tuning of state for memristive devices by ada ptable variation-tolerant algorithm

    Alibart F, Gao L, Hoskins BD, Strukov DB. High prec ision tuning of state for memristive devices by ada ptable variation-tolerant algorithm. Nanotechnology 2012;23:075201. doi: 10.1088/0957-4484/23/7/075201

    2012 doi

  21. [29]

    High precision analogue memristor state tuning

    Berdan R, Prodromakis T, Toumazou C. High precision analogue memristor state tuning. Electron Lett 201 2;48:1105–7. doi: 10.1049/el.2012.2295

    2012

  22. [30]

    Dynamical attractors of memr istors and their networks

    Pershin YV, Slipko VA. Dynamical attractors of memr istors and their networks. EPL (Europhysics Lett 20 19;125:20002. doi: 10.1209/0295-5075/125/20002

    doi

  23. [31]

    Bifurcation analysis of a Ta O memristor model

    Pershin YV, Slipko VA. Bifurcation analysis of a Ta O memristor model. J Phys D Appl Phys 2019;52:50530 4. doi: 10.1088/1361-6463/ab4537

    2019 doi

  24. [32]

    Appl Phys A 2011;102:885–9

    Driscoll T, Pershin Y V, Basov DN, memristor Di Ven tra MChaotic. Appl Phys A 2011;102:885–9. doi: 10.1007/s00339-011-6318- z

    2011 doi

  25. [33]

    Ca- pacitance effects can make memristo r chaotic

    Guseinov D, Matyushkin I, Chernyaev N, Mikhaylov A, Pershin Y. Ca- pacitance effects can make memristo r chaotic. Chaos, Solitons & Fractals 2021;142:1106 99. doi: 10.1016/j.chaos.2021.110699

    2021

  26. [35]

    No ise-induced effects in nonlinear relaxation of con- densed matter systems

    Spagnolo B, Valenti D, Guarcello C, Carollo A, Pers ano Adorno D, Spezia S, Pizzolato N, Di Paola B. No ise-induced effects in nonlinear relaxation of con- densed matter systems. Chaos, Solitons & Fractals 2015;81:412–24. doi: 10. 1016/j.chaos.2015.07.023

    2015

  27. [36]

    Nonlinear Relaxation Pheno mena in Metastable Condensed Matter Systems

    Spagnolo B, Guarcello C, Magazzù L, Carollo A, Pers ano Adorno D, Valenti D. Nonlinear Relaxation Pheno mena in Metastable Condensed Matter Systems. Entrop y 2017;19:20. doi: 10.3390/e19010020

    2017 doi

  28. [37]

    Stabil ization of quantum metastable states by dissipation

    Valenti D, Magazzù L, Caldara P, Spagnolo B. Stabil ization of quantum metastable states by dissipation . Phys Rev B 2015;91:235412. doi: 10.1103/ PhysRevB.91.235412

    2015

  29. [38]

    Verhulst model with Lévy noi se excitation

    Dubkov AA, Spagnolo B. Verhulst model with Lévy noi se excitation. Eur. Phys. J. B 2008;65:361–7. doi: 10.1140/epjb/e2008-00337- 0

    2008 doi

  30. [39]

    Design of a Lambda system for p opulation transfer in superconducting nanocircuits

    Falci G, La Cognata A, Berritta M, D’Arrigo A, Pala dino E, Spagnolo B. Design of a Lambda system for p opulation transfer in superconducting nanocircuits. Phys. Rev. B 2013;87:214515. doi: 10.1103/PhysRevB.87.214515

    2013 doi

  31. [40]

    Volatility Effects on the Es cape Time in Financial Market Models

    Spagnolo B, Valenti D. Volatility Effects on the Es cape Time in Financial Market Models. Int J Bifurc Chaos 2008;18:2775–86. doi: 10.1142/S0218127408022007

    2008 doi

  32. [41]

    A stochastic inter- specific competition mo del to predict the behaviour of Listeria monocytoge nes in the fermentation process of a traditional Sicilian sala mi

    Giuffrida A, Valenti D, Ziino G, Spagnolo B, Panebi anco A. A stochastic inter- specific competition mo del to predict the behaviour of Listeria monocytoge nes in the fermentation process of a traditional Sicilian sala mi. European Food Re- search and Technology 2009;22 8:767–...

    2009 doi

  33. [42]

    Spatio-temporal behaviour of the deep chlorophyll maximum in Mediterranean Sea: Development of a stochastic mode l for pico- phytoplankton dynamics

    Denaro G, Valenti D, La Cognata A, Spagnolo B, Bona nno A, Basilone W, Maz- zola S, Zgozi S, Aronica S. Spatio-temporal behaviour of the deep chlorophyll maximum in Mediterranean Sea: Development of a stochastic mode l for pico- phytoplankton dynamics. Ecological Complexity 20...

    2013 doi

  34. [43]

    Lifetime of metas table states and suppression of noise in Interdis- ciplinary Physical Models

    Spagnolo B, Dubkov AA, Pankratov AL, Pankratova EV, Fiasconaro A, Ochab— Marcinek A. Lifetime of metas table states and suppression of noise in Interdis- ciplinary Physical Models. Acta Phys Pol B 2007;38:1925–50

    2007

  35. [44]

    Enhancement of st ability in randomly switching potential with metast able state

    Spagnolo B, Dubkov AA, Agudov NV. Enhancement of st ability in randomly switching potential with metast able state. The European Physical Journal B 2004;40 :273–81. doi: 10.1140/epjb/e2004-00268-8

    2004 doi

  36. [45]

    Rev Mod Phys 1998;70:223–87

    Gammaitoni L, Hänggi P, Jung P, resonance Marcheson i FStochastic. Rev Mod Phys 1998;70:223–87. doi: 10.1103/RevModPhys.70.223

    1998 doi

  37. [46]

    Resonant activation over a f luctuating barrier

    Doering CR, Gadoua JC. Resonant activation over a f luctuating barrier. Phys Rev Lett 1992;69:2318–21. doi: 10.1103/PhysRevLett.69.2318

    1992 doi

  38. [47]

    Resonant activation in polymer translocation: new insights into the escape dynamics of molecules driven by an oscillating field

    Pizzolato N, Fiasconaro A, Persano Adorno D, Spagno lo B. Resonant activation in polymer translocation: new insights into the escape dynamics of molecules driven by an oscillating field. Physical Biology 2010;7:034001. doi: 10.1088/ 1478-3975/7/3/034001

    2010

  39. [48]

    Effe cts of Lévy noise on the dy- namics of sine-Gordon solitons in long Josephson junctions

    Guarcello C, Valenti D, Carollo A, Spagnolo B. Effe cts of Lévy noise on the dy- namics of sine-Gordon solitons in long Josephson junctions. Journal of St a- tistical Mechanics: Theory and Experiment 2016;2016:054012. doi: 10.1088/ 1742-5468/2016/05/054012

    2016

  40. [49]

    Stochastic memory: Memory enhancement due to noise

    Stotland A, Di Ventra M. Stochastic memory: Memory enhancement due to noise. Phys Rev E 2012;85:011116 . doi: 10.1103/PhysRevE.85.011116

    2012 doi

  41. [50]

    On the benefici al role of noise in resistive switching

    Patterson GA, Fierens PI, Grosz DF. On the benefici al role of noise in resistive switching. Appl Phys Lett 2013;103:074102. doi: 10.1063/1.4819018

    2013 doi

  42. [51]

    Experimental Inves- tigation of Memristance E nhancement

    Ntinas V, Rubio A, Sirakoulis GC, Rodriguez R, Nafr ia M. Experimental Inves- tigation of Memristance E nhancement. IEEE/ACM Int. Symp. Nanoscale Archit. 2 019;1:1–2. doi: 10.1109/NANOARCH47378.2019.181299

    2019

  43. [52]

    Power-efficient Noise-Ind uced Reduction of ReRAM Cell’s Temporal Variability Effects

    Ntinas V, Rubio A, Sirakoulis GC, Aguilera ES, Pedr o M, Crespo-Yepes A, et al. Power-efficient Noise-Ind uced Reduction of ReRAM Cell’s Temporal Variability Effects. IEEE Trans Circuits Syst II Express Briefs 2020;7747 1–1 . doi: 10.1109/ TCSII.2020.3026950

    2020

  44. [53]

    Noise-induced resistive switching in a memristor based on ZrO 2 (Y)/Ta 2 O5 stack

    Filatov DO, Vrzheshch DV, Tabakov OV, Novikov AS, B elov AI, Antonov IN, et al. Noise-induced resistive switching in a memristor based on ZrO 2 (Y)/Ta 2 O5 stack. J Stat Mech Theory Exp 2019;2019:124026. doi: 10.1088/1742-5468/ ab5704

    2019 doi

  45. [54]

    Time series statis- tical analysis: A powerfu l tool to evaluate the variability of resistive swi tching memories

    Roldán JB, Alonso FJ, Aguilera AM, Maldonado D, Lan za M. Time series statis- tical analysis: A powerfu l tool to evaluate the variability of resistive swi tching memories. J Appl Phys 2019;125:174504. doi: 10.1063/1.5079409

    2019 doi

  46. [55]

    Expe rimental study of a non- linear system in the prese nce of noise: The stochastic resonance

    Lanzara E, Mantegna RN, Spagnolo B, Zangara R. Expe rimental study of a non- linear system in the prese nce of noise: The stochastic resonance. Am J Phys 1 997;65:341–9. doi: 10.1119/1.18520

    doi

  47. [56]

    Phys Rev Lett 1988 ;60:2626–9

    McNamara B, Wiesenfeld K, Roy R. Phys Rev Lett 1988 ;60:2626–9. doi: 10.1103/ PhysRevLett.60.2626

    1988

  48. [57]

    Stochastic Resonance in a Tunnel Diode in the Pres- ence of White or Colored Noise

    Mantegna RN, Spagnolo B. Stochastic Resonance in a Tunnel Diode in the Pres- ence of White or Colored Noise. Nuovo Cimento D 1995;17:873–81. doi: 10. 1007/BF02451845

    1995

  49. [58]

    Linear and No nlinear Experimen- tal Regimes of Stochastic Resona nce

    Mantegna RN, Spagnolo B, Trapanese M. Linear and No nlinear Experimen- tal Regimes of Stochastic Resona nce. Phys Rev E 2001;63:011101. doi: 10.1103/ PhysRevE.63.011101

    2001

  50. [59]

    Andronov A et al. Proc. II Workshop NATO SfP973799 Semiconductors (Nizhnii Novgorod, Russia, 2002) 200 2: 38

    2002

  51. [60]

    Transient b imodality of nonequi- librium states in monostable systems with noise

    Agudov NV, Devyataykin RV, Malakhov AN. Transient b imodality of nonequi- librium states in monostable systems with noise. Radiophys Quantum Electron 1999 ;42:902–

    1999

  52. [61]

    doi: 10.1007/BF02677104

    doi

  53. [62]

    Resistive Switching and C harge Transport in Laser-Fabricated Graphene Oxide Memristors: A Time Series and Quantum Point Contact Modeling Appr oach

    Rodriguez N, Maldonado D, Romero FJ, Alonso FJ, Agu ilera AM, Godoy A, et al. Resistive Switching and C harge Transport in Laser-Fabricated Graphene Oxide Memristors: A Time Series and Quantum Point Contact Modeling Appr oach. Materials (Basel) 2019;12:3734. doi: 10.3390/ma12223734

    2019 doi

  54. [63]

    A comprehensive analysis o n progressive reset transitions in RRAMs

    Villena MA, Roldán JB, Jimenez-Molinos F, Suñé J, L ong S, Miranda E, et al. A comprehensive analysis o n progressive reset transitions in RRAMs. J Phys D Appl Phys 2014;47:205102. doi: 10.1088/0022-3727/47/20/205102

    2014 doi

  55. [64]

    Introduction to Time Series and Forecasting 2016;92

    Brockwell PJ, Davis RA. Introduction to Time Series and Forecasting 2016;92. doi: 10.1007/978- 3- 319-29854- 2

    2016 doi

  56. [65]

    Time Series Analysis and For ecasting by Example

    Bisgaard S, Kulahci M. Time Series Analysis and For ecasting by Example. Wiley; 2011. doi: 101002/9781118056943

    2011

  57. [66]

    Stochastic resonance s ubject to multiplicative and additive noise: The in fluence of potential asymmetries

    Qiao Z, Lei Y, Lin J, Niu S. Stochastic resonance s ubject to multiplicative and additive noise: The in fluence of potential asymmetries. Phys Rev E 2016;9 4:052214. doi: 10.1103/PhysRevE.94.052214

    2016 doi

  58. [67]

    The Fourier s ignatures of memristive hysteresis 2020

    Pershin Y V., Chien C-C, Di Ventra M. The Fourier s ignatures of memristive hysteresis 2020. arXiv:2010 .01313

    2020

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.