DETAILED ACTION
Claims 1-3 and 5-6 are presented for examination.
Claims 1-3 and 5-6 have been amended.
This office action is in response to the amendment submitted on 30-MAR-2026.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments – 35 USC 101
The Applicant/Arguments Remarks, Applicant argues the amended claims have overcome the rejection under 35 USC 101.
Applicant's arguments have been fully considered and they are persuasive. Rejection under 35 USC 101 is withdrawn.
Response to Arguments – 35 USC 103
Applicant’s arguments with respect to the 103 rejections have been considered, but are moot in view of the new ground(s) of rejection provided below.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1 and 6 are rejected under 35 U.S.C. 103 as being unpatentable over Shrauwen et al. (Improving reservoirs using intrinsic plasticity) in view of Kanao et al. (Reservoir Computing on Spin-Torque Oscillator Array) and further view of Romera (Vowel recognition with four coupled spin-torque nano-oscillators)
Regarding Claim 1, Shrauwen teaches A control method for a physical reservoir device including a plurality of elements, the method comprising:
(a) performing pre-training such that a Kullback-Leibler divergence between (i) an ideal probability distribution of an output of the reservoir device, the ideal probability distribution being derived from a device model based on characteristics of the plurality of elements, and (ii) an actual probability distribution of the output of the reservoir device is equal to or less than a prescribed threshold; (pg. 7, “For pre-training a reservoir, the IP rule is applied with a learning rate of 0.0005 for 100 000 time steps (equal to 10 epochs when we use 10 time series, used for the cross validation, each consisting of 1000 time steps). To check whether IP has had sufficient time to adapt after this time, we verified that a and b had converged to small regions and compared the expected probability density with the one estimated from the reservoir’s output” Pg. 3-4, “We express the amount the empirical output distribution differs from the desired ME distribution using the Kullback–Leibler divergence … where p~(y) is the actual probability density of the neuron’s output activity and p(y) the desired probability density function … To extend the formalism to include the second moment, we need to minimize the Kullback–Leibler divergence to a desired Gaussian distribution” EN: Minimizing the divergence increases the mutual information between the ideal and output distribution. The divergence criteria for Shrawen is 100,000 time steps. Using a set iterations or a min threshold is a common design choice for ML training. The reservoir output is the aggregate array level output).
(b) determining a non-uniform parameter distribution for parameters that define variations of the respective ones of the plurality of elements in the device model; (Pg. 3, “We define intrinsic parameters for these non-linearities by adding a gain a and a bias b: f gen(x)=f (ax + b)” Pg. 4 describes how gain and bias are updated per neuron/element they define the variations per element. They are updated per neuron element the totality of them is non-uniform parameter distribution)
However, Shrawen is not relied upon for:
(c) converting the determined parameter distribution into a distribution of
(d) physically applying, to each of the plurality of elements, an element- specific control input to realize a physical operating characteristic based on the distribution of physical characteristics
Kanoa teaches (c) converting the determined parameter distribution into a distribution of (Pg. 1-2, “For reservoir computing, we utilize the dynamics of the oscillation powers and phases (or frequencies) of the STOs [13–16]. To calculate this dynamics, we use a corresponding nonlinear oscillator model [12, 32], which can capture the essential features of the dynamics of the STOs, including the synchronization [33], irrespective of the detailed magnetization configuration of the STOs. The nonlinear oscillator model can be derived from the LLGS equation under certain assumptions. (See Appendix B for details.) The powers and frequencies of the STOs usually vary slowly compared with the oscillation itself [12, 25], and fast oscillation components can be eliminated in the nonlinear oscillator model …
PNG
media_image1.png
81
446
media_image1.png
Greyscale
where pi0, Nf , Γp, and ξi(t) are the stationary oscillation power, the nonlinear frequency shift, the rate of damping of power, and the modulation by an input signal, respectively, in units of frequency [12]. δωi(pi) and Γi(pi) describe the properties of a single STO, and δωi(pi0) = Γi(pi0) = 0 holds for its stationary oscillation state for a constant current. We add the input signal to the current, which, via the spin-transfer torque, leads to a modulation of Γi(pi)… Differences between the STOs are modeled by taking pi0 as a Gaussian random variable with a mean of p0.” Kanao converts the parameter distribution pi0 through equation 2 to the characteristic distribution).
Shrauwen and Kanao are analogous art because they are from the same field of endeavor in reservoir computing. Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art, to combine Shrauwen and Kanao to incorporate Kanao’s STO oscillators as a reservoir computing device model.
Romera teaches (c) a distribution ofphysical characteristics of the plurality of elements (Pg. 21, “In this equation, 𝜇 = 0.1 𝑚𝐴 is the learning rate of our algorithm. At each step, the applied dc current through each oscillator can be modified only by ±𝜇. Here 𝑠𝑔𝑛[ (𝜕𝜔𝑘 𝜕𝐼 )𝐼=𝐼𝑘 ] represents the sign of the frequency evolution versus injected dc current of the kth-oscillator at the value of current 𝐼𝑘. For this, the frequency – current dependence of each independent oscillator has been previously characterized.” EN: Romera convert the distribution into a physical hardware characteristic/dc current)
(d) physically applying, to each of the plurality of elements, an element- specific control input to realize a physical operating characteristic based on the distribution of physical characteristics (Pg. 2, “we show that the outstanding tunability of spintronic nano-oscillators, i.e. the possibility to widely and accurately control their frequency through electrical current and magnetic field, can solve this challenge. We successfully train a hardware network of four spin-torque nano-oscillators to recognize spoken vowels by tuning their frequencies according to an automatic real-time learning rule.” Pg. 18, “Each spectrum recorded with the spectrum analyzer is sent to the computer, where it is analyzed by a program in real time. The information we use as input to this program is: (i) the value of the two frequencies of the external microwave signals (fA, fB) and (ii) the oscillator frequencies at each dc current values in the absence of external microwave signals (f10, f20, f30, f40). The output data that we extract from each spectrum analysis are the four values of the oscillator frequencies in the presence of microwave inputs. Then, another program takes these oscillator frequencies to calculate the synchronization states and check if the applied vowel was properly recognized”)
PNG
media_image2.png
369
681
media_image2.png
Greyscale
Shrauwen, Kanao and Ramera are analogous art because they are from the same field of endeavor in reservoir computing. Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art, to combine Shrauwen, Kanao and Ramera to apply the training from Shrauwen and modeling from Kanao to a physical hardware system as provided by Ramera.
Regarding claim 6, Shrauwen in view of Kanao and further in view of Ramera teaches the control method according to claim 1. Ramera teaches wherein the plurality of elements are spin-torque oscillators arranged in a spin-torque oscillator array (Please see Extended data Figure 1, where the STO are arranged in a STO array configuration)
Claims 2 is rejected under 35 U.S.C. 103 as being unpatentable over Shrauwen et al. (Improving reservoirs using intrinsic plasticity) in view of Kanao et al. (Reservoir Computing on Spin-Torque Oscillator Array) in further view of Ramera and further in view of Coulombe et al. (Computing with networks of nonlinear mechanical oscillators)
Regarding claim 2, Shrauwen in view of Kanao and further in view of Ramera teaches the method of claim 1. Coulombe further teaches the parameter setting model according to claim 1, wherein the device model is a model based on spring vibration and is expressed by
PNG
media_image3.png
109
561
media_image3.png
Greyscale
(Coulombe’s equation is explicit. Pg. 2, Eq 2,
PNG
media_image4.png
73
507
media_image4.png
Greyscale
Pg. 2, “where dots denote derivatives with respect to time, i = 1, . . ., N”).
PNG
media_image5.png
278
567
media_image5.png
Greyscale
Shrauwen, Kanao, Ramera and Coulombe are analogous art because they are from the same field of endeavor in reservoir computing. Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art, to combine Shrauwen, Kanao, Ramera and Coulombe to incorporate use Coulombe’s explicit STO equation. “We describe a mechanical device, which operates in a manner similar to artificial neural networks, to solve efficiently two difficult benchmark problems (computing the parity of a bit stream, and classifying spoken words)” (Coulombe, Abstract).
Claims 3 is rejected under 35 U.S.C. 103 as being unpatentable over Shrauwen et al. (Improving reservoirs using intrinsic plasticity) in view of Kanao et al. (Reservoir Computing on Spin-Torque Oscillator Array) in view of Ramera and further in view of Slavin et al. (IDS Ref: Nonlinear Auto-Oscillator Theory of Microwave Generation by Spin-Polarized Current)
Regarding claim 3, Shrauwen in view of Kanao and further in view of Ramera teaches the method of claim 1. Slavin teaches wherein the device model is a model based on a generalized nonlinear vibrator model and is expressed by
PNG
media_image6.png
66
322
media_image6.png
Greyscale
(While Kanao teach the same equation in an alternate form, Slavin is explicit, Eq. 4:
PNG
media_image7.png
52
428
media_image7.png
Greyscale
And pg. 6, “The majority of auto-oscillators of this type, regardless of a particular physical realization, can be described by the same nonlinear oscillator model”).
Shrauwen, Kanao, Ramera and Slavin are analogous art because they are from the same field of endeavor in STO modeling. Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art, to combine Shrauwen, Kanao, Ramera and Slavin to use Slaven’s more explicit STO equation. “In this section we demonstrate explicitly how the nonlinear oscillator model (5) can be derived for the classical electrical auto-oscillatory circuit (van der Pol oscillator) and for the spin-torque oscillator in the simplest possible geometry (normally magnetized magnetic nano-pillar)” (Slavin, Pg. 6).
Claims 5 is rejected under 35 U.S.C. 103 as being unpatentable over Shrauwen et al. (Improving reservoirs using intrinsic plasticity) in view of Kanao et al. (Reservoir Computing on Spin-Torque Oscillator Array) in further view of Ramera and further in view of Harney et al. (US20180332390A1)
Regarding claim 5, Harney teaches the control method according to claim 1, wherein the plurality of elements are MEMES microphones arranged in a MEMS microphone array ([0010] “n further non-limiting implementations, the disclosed subject matter provides exemplary MEMS microphones comprising an ASIC having a positive bias voltage generator or a negative bias voltage generator that facilitates providing output signals having the same polarity as the incident acoustic waves or sound pressure. In further non-limiting implementations, the disclosed subject matter provides exemplary MEMS microphones comprising an ASIC having a single bias voltage generator and a differential amplifier arrangement that facilitates providing output signals having the same polarity as the incident acoustic waves or sound pressure as well as improved opportunities for rejection of common mode interference.” EN: both Kanao and Ramera provide the array structure for the reservoir elements)
Shrauwen, Kanao, Ramera and Harney are analogous art because they are from the same field of endeavor in STO modeling. Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art, to combine Shrauwen, Kanao, Ramera and Harney to use Shrauwen’s model to MEMES microphones as disclosed in Harney.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AMIR DARWISH whose telephone number is (571)272-4779. The examiner can normally be reached 7:30-5:30 M-Thurs.
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Torrejon et at (Neuromorphic computing with nanoscale spintronic oscillators)
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Lewis Bullock can be reached on 571-272-3759. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/A.E.D./Examiner, Art Unit 2199
/LEWIS A BULLOCK JR/Supervisory Patent Examiner, Art Unit 2199