Detailed Office Action
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 .
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
Response to Arguments
Applicant’s arguments with respect to claims 1-11 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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 of this title, 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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-6 and 10-11
Claims 1-6 and 10-11 are rejected under 35 U.S.C. 103 as being unpatentable over Williamson et al. (Reprogrammable Electro-Optic Nonlinear Activation Functions for Optical Neural Networks, IEE Journal of Selected Topics in Quantum Electronics, V. 26, N. 1, 2020; “Williamson”) in view of Bandyopadhyay et al. (Single chip photonic deep neural network with accelerated training. arXiv:2208.01623 (2022); “Bandyopadhyay”).
Regarding claim 1, Williamson discloses in figures 1 and 5, and related figures and text, for example, Selected Text, embodiments of optical devices comprising a waveguide receiving an input signal encoded with vectors, a plurality of optical-to-electrical (O/E) converters, and a plurality of optical modulators that are optically coupled in series f1 to Fl to the waveguide such that the output optical signal is encoded with a product of the first vector and the plurality of vectors. See below, Williamson, figures 1 and 5, and related figures and text, for example, Selected Text.
Disclosing “a function which accepts an input vector, x0 and returns an output vector;” “a layer-by-layer fashion, with each layer consisting of a linear matrix-vector multiplication followed by the application of an element-wise nonlinear function, or activation, on the result nonlinear activation functions in optical neural networks;” “optical-to-optical nonlinearity operates by converting a small portion of the input optical signal into an analog electric signal, which is used to intensity -modulate the original optical signal …; “… the information being processed by the network, xi, is encoded into the modal amplitudes of the waveguides feeding the device and the matrix-vector multiplications are accomplished using meshes of integrated optical interferometer;” “implements an optical-to-optical nonlinearity by converting a small portion of the optical input power into an electrical voltage[;]The remaining portion of the original optical signal is phase and amplitude-modulated by this voltage as it passes through an interferometer.” and
Disclosing “a feedforward neural network of L layers. Each layer consists of a ˆWi block representing a linear matrix which multiplies vector inputs xi−1. The fi block in each layer represents an element-wise nonlinear activation function operating on vectors zi to produce outputs xi.”
Williamson, figures 1 and 5, and related figures and text, for example, Selected Text.
Further regarding claim 1, Bandyopadhyay discloses in figures 1-3, and related figures and text, for example, Selected Text, embodiments of optical devices that explicitly maps input vectors: “Light is fiber coupled into a single input on the chip and fanned out to the six channels of the transmitter (i). Each channel encodes the amplitude and phase of one element of the input x(j) into the optical field a(1) (j) with a Mach-Zehnder modulator and an external phase shifter. The coherent matrix multiplication unit (ii), consisting of a Mach-Zehnder interferometer mesh, implements the linear transformation b(n)…Programmable nonlinear optical function units (iii) realize activation functions a(n+1)… by tapping off part of the signal to a photodiode …Transimpedance amplifiers convert the output photocurrents to voltages, …which drives a cavity off-resonance by injecting carriers into the waveguide…”). See below, Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text.
Consequently, in light of Bandyopadhyah’s disclosure of mapping input vectors, it would have been obvious to one of ordinary skill in the art to modify Williamson’s embodiments to disclose an optical device comprising: a waveguide to receive an input optical signal encoded with a first vector and output an output optical signal; a plurality of optical-to-electrical (O/E) converters, each O/E converter of the plurality of O/E converters to receive a respective optical signal of a plurality of optical signals encoded with a respective vector of a plurality of vectors and generate a respective electrical signal of a plurality of electrical signals based on the respective optical signal; a plurality of optical modulators optically coupled to the waveguide in series, each optical modulator of the plurality of optical modulators electrically coupled to a respective one of the plurality of O/E converters, and each optical modulator to modulate an intensity of the input optical signal on the waveguide based on the respective electrical signal generated by the respective one of the plurality of O/E converters electrically coupled to the optical modulator; and wherein the output optical signal is encoded with a product of the first vector and the plurality of vectors; Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text; because the resultant configurations and methods would facilitate designing, fabricating, and deploying coherent optical devices that perform ‘both inference and in situ training.’ Bandyopadhyay, Conclusion.
Williamson – Figures 1 and 5, and Selected Text
PNG
media_image1.png
557
397
media_image1.png
Greyscale
PNG
media_image2.png
242
402
media_image2.png
Greyscale
PNG
media_image3.png
56
405
media_image3.png
Greyscale
Abstract—We introduce an electro-optic hardware platform for nonlinear activation functions in optical neural networks. The optical-to-optical nonlinearity operates by converting a small portion of the input optical signal into an analog electric signal, which is used to intensity -modulate the original optical signal with no reduction in processing speed. Our scheme allows for complete nonlinear ON–OFF contrast in transmission at relatively low optical power thresholds and eliminates the requirement of having additional optical sources between each of the layers of the network Moreover, the activation function is reconfigurable via electrical bias, allowing it to be programmed or trained to synthesize a variety of nonlinear responses. Using numerical simulations, we demonstrate that this activation function significantly improves the expressiveness of optical neural networks, allowing them to perform well on two benchmark machine learning tasks: learning a multi-input exclusive-OR (XOR) logic function and classification of images of handwritten numbers from the MNIST dataset. The addition of the nonlinear activation function improves test accuracy …
…machine learning applications, such as artificial neural networks (ANNs) …first optical neural networks (ONNs) …
As outlined in Fig. 1(a), an ANN is a function which accepts an input vector, x0 and returns an output vector, xL. This is accomplished in a layer-by-layer fashion, with each layer consisting of a linear matrix-vector multiplication followed by the application of an element-wise nonlinear function, or activation, on the result. …For a given ANN, a loss function is defined to quantify the difference between the target output and output predicted by the network. During training, this loss function is minimized with respect to tunable degrees of freedom, namely the elements of the weight matrix ˆWi within each layer. In general, although less common, it is also possible to train the parameters of the activation functions…
linear operations are implemented using an integrated optical circuit … the information being processed by the network, xi, is encoded into the modal amplitudes of the waveguides feeding the device and the matrix-vector multiplications are accomplished using meshes of integrated optical interferometers.
an electro-optic architecture for synthesizing optical-to-optical nonlinearities … by measuring a small portion of the incoming optical signal power and using electro-optic modulators to modulate the original optical signal, without any reduction in operating bandwidth or computational speed. … the application of our architecture as an element-wise activation in a feedforward ONN
realizing the activation function, fi(·), on-chip with a hybrid electro-optic circuit feeding an inteferometer. In Fig. 1(b), we show how this activation scheme fits into a single layer of an ONN …
III. NONLINEAR ACTIVATION FUNCTION ARCHITECTURE In this section, we describe our proposed nonlinear activation function architecture for optical neural networks, which implements an optical-to-optical nonlinearity by converting a small portion of the optical input power into an electrical voltage. The remaining portion of the original optical signal is phase and amplitude-modulated by this voltage as it passes through an interferometer. For an input signal with amplitude z, the resulting nonlinear optical activation function, f(z), is a result of the responses of the interferometer under modulation as well as the components in the electrical signal pathway.
A schematic of the architecture is shown in Fig. 1(c), where black and blue lines represent optical waveguides and electrical signal pathways, respectively. The input signal first enters a directional coupler which routes a portion, α, of the input optical power to a photodetector. The photodetector is the first element of an optical-to-electrical conversion circuit, which is a standard component of high-speed optical receivers for converting an optical intensity into a voltage. In this work, we assume a normalization of the optical signal such that the total power in the input signal is given by |z|2. The optical-to-electrical conversion process consists of the photodetector producing an electrical current, Ipd = R · α|z|2, where Ris the photodetector responsivity, and a transimpedance amplifying stage, characterized by a gain G, converting this current into a voltage VG = G · R · α|z|2. The output voltage of the optical-to-electrical conversion circuit then passes through a nonlinear signal conditioner with a transfer function, H(·). This component allows for the application of additional nonlinear functions to transform the voltage signal. Finally, the conditioned voltage signal, H(VG) is combined with a static bias voltage, Vb to induce a phase shift … for the optical signal routed through the lower port of the directional coupler. The parameter Vπ represents the voltage required to induce a phase shift of π in the phase modulator. This phase shift, defined by Eq. 2, is a nonlinear self-phase modulation because it depends on the input signal intensity.
An optical delay line between the directional coupler and the Mach-Zehnder interferometer (MZI) is used to match the signal propagation delays in the optical and electrical pathways. This ensures that the nonlinear self-phase modulation defined by Eq. (2) is applied at the same time that the optical signal which generated it passes through the phase modulator….
The nonlinear self-phase modulation achieved by the electric circuit is converted into a nonlinear amplitude response by the MZI, …a larger input optical signal amplitude causes either more or less power to be diverted away from the output port, resulting in a nonlinear self-intensity modulation. Combining …
A benefit of having electrical control over the activation response is that, in principle, its electrical bias can be connected to the same control circuitry which programs the linear interferometer meshes. In doing so, a single ONN hardware unit can then be reprogrammed to synthesize many different activation function responses. This opens up the possibility of heuristically selecting an activation function response, or directly optimizing the activation bias using a training algorithm. This realization of a flexible optical-to-optical nonlinearity can allow ONNs to be applied to much broader classes of machine learning tasks.
…the electro-optic scheme has several degrees of freedom which allow it to potentially achieve a stronger nonlinear response …
A. Exclusive-OR Logic Function An exclusive-OR (XOR) is a logic function which takes two inputs and produces a single output. The output is high if only one of the two inputs is high, and low for all other possible input combinations. In this example, we consider a multi-input XOR which takes N input values, given by x1 . . . xN, and produces a single output value, y. The input-output relationship of the multi-input XOR function is a generalization of the two-input XOR. For example, defining logical high and low values as 1 and 0, respectively, a four-input XOR has an output table indicated the desired values in Fig. 5(b). We select this task for the ONN to learn because it requires a non-trivial level of nonlinearity, meaning that it could not be implemented in an ONN consisting of only linear interferometer meshes.
The architecture of the ONN used to learn the XOR is shown schematically in Fig. 5(a). The network consists of L layers, with each layer constructed from an N × N unitary interferometer mesh followed by an array of N parallel electro-optic activation functions, with each element corresponding to the circuit in Fig. 1(c). After the final layer, the lower N − 1 outputs are dropped to produce a single output value which corresponds to y….
…Rather than using all-optical nonlinearities, our activation architecture uses intermediate signal pathways in the electrical domain which are accessed via photodetectors and phase modulators. Specifically, a small portion of the optical input power is tapped out which undergoes analog processing before modulating the remaining portion of the same optical signal. Whereas all-optical nonlinearities have largely fixed responses, a benefit of the electro-optic approach demonstrated here is that signal amplification in the electronic domain can overcome the need for high optical signal powers to achieve a significantly lower activation threshold…
…the majority of the signal power remains in the optical domain. There is no need to have a new optical source at each nonlinear layer of the network, …each activation function in our proposed scheme is a standalone analog circuit and therefore can be applied in parallel….this circuit could be of broader interest for optical computing as well as microwave photonic signal processing applications.
Bandyopadhyay, Figures 1-3, and Selected Text
PNG
media_image4.png
543
605
media_image4.png
Greyscale
PNG
media_image5.png
680
640
media_image5.png
Greyscale
PNG
media_image6.png
460
604
media_image6.png
Greyscale
As deep neural networks (DNNs) revolutionize machine learning [1–6], energy consumption and throughput are emerging as fundamental limitations of CMOS electronics. This has motivated a search for new hardware architectures optimized for artificial intelligence, such as electronic systolic arrays [7], memristor crossbar arrays [8], and optical accelerators. Optical systems can perform linear matrix operations at exceptionally high rate and efficiency [9], motivating recent demonstrations of low latency linear algebra [10–14] and optical energy consumption [15, 16] below a photon per multiply-accumulate operation. However, demonstrating systems that co-integrate both linear and nonlinear processing units in a single chip remains a central challenge. Here we introduce such a system in a scalable photonic integrated circuit (PIC), enabled by several key advances: (i) high-bandwidth and low-power programmable nonlinear optical function units (NOFUs); (ii) coherent matrix multiplication units (CMXUs); and (iii) in situ training with optical acceleration. We experimentally demonstrate this fully-integrated coherent optical neural network (FICONN) architecture for a 3-layer DNN comprising 12 NOFUs and three CMXUs operating in the telecom C-band. Using in situ training on a vowel classification task, the FICONN achieves 92.7% accuracy on a test set, which is identical to the accuracy obtained on a digital computer with the same number of weights. This work lends experimental evidence to theoretical proposals for in situ training, unlocking orders of magnitude improvements in the throughput of training data. Moreover, the FICONN opens the path to inference at nanosecond latency and femtojoule per operation energy efficiency.
Figure 1 shows how this PIC architecture allows us to realize a fully integrated coherent optical neural network (FICONN) through the following stages: (i) the transmitter (TX) maps input vectors x(j) to an optical field vector a(1) (j) by splitting an input laser field into MZI modulators m = 1, 2, ..., 6, each of which encode one element of x(j) into the amplitude Am and phase _m of the transverse electric field component a(1) (j ),m = Amei!t+_m ; (ii) the coherent matrix multiplication unit (CMXU), consisting of a MZI mesh [10, 26, 30], transforms a(1) ! b(1) = U(1)a(1) through passive optical interference; and (iii) the programmable nonlinear optical function unit (NOFU) applies the activation function to yield the input to the next layer, a(2) = f (b(1)). Following the input layer, the PIC directly transmits the optically-encoded signal into a hidden layer, composed of another CMXU and six NOFUs, that implements the transformation a(3) = f (U(2)a(2)). The final layer U(3), implemented Inference therefore proceeds entirely in the optical domain without photodiode readout, amplifiers, or digitization between layers.
An integrated coherent receiver (ICR), shown in Figure 1(iv), reads out the amplitude and phase of the DNN output by homodyning each element of the vector b(3) with a common local oscillator field ELO. The DNN output is read out by transimpedance amplifiers that convert the photocurrent vector iPD to a voltage vector VPD. VPD is digitized and then normalized by the sum of voltages measured across all channels P VPD to yield a quasi-probability distribution Vnorm for a classification task. Each sample x(i) is assigned the label corresponding to the highest probability, i.e. argmax(Vnorm).
Figure 2e shows the integrated coherent receiver (ICR), which measures the amplitude and phase of the output signal b(3) of the DNN. Each channel, as shown in Figure 2f, interferes the signal field with the LO using a 50-50 multimode interferometer (MMI) and measures the outputs with a pair of balanced detectors. A phase shifter is used to select the quadrature being read out.
The programmable nonlinear optical function unit (NOFU) is shown in Figure 3. To realize a programmable coherent optical activation function, we developed the resonant electro-optical nonlinearity shown schematically in Figure 1iii). This device directs a fraction _ of the incident optical power |b|2 into a photodiode by programming the phase shift _ in an MZI. The photodiode is electrically connected to a pn-doped resonant microring modulator, and resonance by either injecting (or depleting) carriers from the waveguide. The remainder of the incident signal field passes into the microring resonator; the nonlinear modulation of the electric field b by the cavity, which is dependent on the incident optical power |b|2, results in a coherent non linear optical function for DNNs. Setting the detuning of the cavity and the fraction of optical power tapped off to the photodiode determines the implemented function.
Figure 3a shows the fabricated device, where the photodiode output is directly connected on chip to the modulator. An integrated heater aligns the microring resonance to the programmed detuning, and an optical delay line placed between the tunable coupler and modulator synchronizes the optical and electrical pulses.
The electrical circuit for the NOFU is shown in Figure 3b. Incident light generates a reverse current in the photodiode; depending on the bias voltage VB, this either injects carriers into the modulator or generates a photovoltage that depletes the modulator of carriers. Figure 3c shows the device response in injection (left) and depletion modes (right). In injection mode, optical power modulates both the loss and phase of the resonator, producing a strong nonlinear response to the incident field b. In depletion mode, we observe nearly a linewidth detuning when the incident light is switched on vs. off, which is induced by the voltage produced by the photodiode.
The NOFU is designed to implement programmable nonlinear activation functions at high speeds with ultra-low energy consumption. This required separately optimizing the cavity parameters, which determine the microring response time, and closely integrating the photodiode and modulator together on the PIC to minimize total device capacitance, and therefore the RC time delay. In injection mode, we found that 75 μA photocurrent was sufficient to detune the resonator by a linewidth. As each NOFU performs the equivalent of two multiplications in digital electronics, over a carrier lifetime of _1 ns this corresponds to an energy consumption of 30 fJ per nonlinear operation (NLOP). Compared to prior approaches [14, 24], the NOFU directly drives the modulator through the photodiode and eliminates the amplifier stage between them. This greatly improves the latency and energy efficiency of the device, as high speed transimpedance amplifiers consume hundreds of milliwatts of power [36]. For our device, incorporating such an amplifier would have increased the power consumption by two orders of magnitude to about 3 pJ/NLOP. Our design, which eliminates intermediate amplifier circuitry and is therefore “receiverless” [37], is not only more energy-efficient, but also eliminates the latency introduced by the amplifier. In Figure 3d, we show several of the activation functions measured on chip. The programmability of the device enables a wide range of nonlinear optical functions to be realized. By tuning the fraction of power tapped off to the photodiode and the relative detuning of the cavity, we can not only program the form of the nonlinear function, but also train it during model optimization.
Regarding claims 2-6 and 10-11, it would have been obvious to one of ordinary skill in the art to modify Williamson in view of Bandyopadhyay, as applied in the rejection of claim 1, to disclose:
2. The optical device of claim 1, wherein the one or more O/E converters are photodetectors. Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text.
3. The optical device of claim 1, wherein each of the one or more optical modulators comprises: a phase shifter electrically coupled to a respective O/E converter; and a Mach-Zehnder interferometer (MZI) optically coupled to the waveguide and to the phase shifter. Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text.
4. The optical device of claim 3, further comprising: one or more linear transformation devices electrically connected to and between the one or more O/E converters and the one or more optical modulators, wherein the one or more linear transformation devices are configured to convert photocurrents generated by the one or more O/E converters to the one or more electrical signals, wherein the phase shifters are configure to apply a phase change that is proportional to the one or more electrical signals. Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text.
5. The optical device of claim 4, wherein the one or more linear transformation devices are one or more resistors. Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text.
6. The optical device of claim 3, wherein the MZI comprises a micro-ring assisted MZI (RAMZI), wherein the phase shifter is disposed on a micro-ring of the RAMZI. Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text.
10. The optical device of claim 1, wherein the one or more O/E converters comprises at least: a first O/E converter to receive a first optical signal encoded with a second vector, and a second O/E converter to receive a second optical signal encoded with a third vector, wherein the output optical signal is encoded with an elementwise product of the first vector and the second and third vectors. Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text.
11. The optical device of claim 9, wherein the one or more optical modulators comprises at least: a first optical modulator optically coupled to the waveguide and electrically coupled to the first O/E converter, and a second optical modulator optically coupled to the waveguide and electrically coupled to the second O/E converter. Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text.
because the resultant configurations and methods would facilitate designing, fabricating, and deploying coherent optical devices that perform ‘both inference and in situ training.’ Bandyopadhyay, Conclusion.
Claims 7 and 8
Claims 7 and 8, as dependent upon claim 6, are rejected under 35 U.S.C. 103 as being unpatentable over Williamson et al. (Reprogrammable Electro-Optic Nonlinear Activation Functions for Optical Neural Networks, IEE Journal of Selected Topics in Quantum Electronics, V. 26, N. 1, 2020; “Williamson”) in view of Bandyopadhyay et al. (Single chip photonic deep neural network with accelerated training. arXiv:2208.01623 (2022); “Bandyopadhyay”), as applied in the rejection of claims 1-6 and 10-11, and further in view of Sacher et al. (Dynamics of microring resonator modulators, Opt. Express 16, 15741-15753 (2008); “Sacher”).
Regarding claims 7 and 8, Sacher discloses embodiments of microring modulators for which, “at higher frequencies, the modulation depth is slightly larger for the over-coupled (σ < a) ring.” Sacher, Numerical Results and Sacher, figure 1, and related text, for example, 2. Time-dependent microring transmission (“The situation when σ = a is referred to as critical coupling. At critical coupling, the wave in the bus waveguide destructively interferes with the wave coupled out of the ring to result in zero transmission [11, 12]. To have complete extinction of the input wave, the modulator must thus operate near the critical coupling condition. Moreover, to use small changes in the index, loss, or coupling to cause large changes in the output intensity, the Q of the resonator must be high (a, σ ≈ 1), so that a circulating wave can, in essence, experience any small changes in device parameters many times before being dissipated.”).
Consequently, it would have been obvious to one of ordinary skill in the art to modify Williamson in view of Bandyopadhyay, as applied in the rejection of claims 1-6 and 10-11, to disclose:
7. The optical device of claim 6, wherein the micro-ring is overcoupled to the MZI. Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text; Sacher, figure 1, and related text.
8. The optical device of claim 7, wherein a coupling coefficient between the micro-ring and the MZI is between 0.89 and 0.94. Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text; Sacher, figure 1, and related text.
because the resultant configurations and methods would facilitate designing, fabricating, and deploying coherent optical devices that perform ‘both inference and in situ training;’ by predictably tailoring intensity modulation characteristics. Sacher, abstract.
Claim 9
Claim 9, as dependent upon claim 3, is rejected under 35 U.S.C. 103 as being unpatentable over Williamson et al. (Reprogrammable Electro-Optic Nonlinear Activation Functions for Optical Neural Networks, IEE Journal of Selected Topics in Quantum Electronics, V. 26, N. 1, 2020; “Williamson”) in view of Bandyopadhyay et al. (Single chip photonic deep neural network with accelerated training. arXiv:2208.01623 (2022); “Bandyopadhyay”), as applied in the rejection of claims 1-6 and 10-11, and further in view of Popovic et al. (2010/0209038; “Popovic”).
Regarding claim 9, Popovic disclosures embodiments of microring resonators and Mach-Zehnder structures configured such that ‘when non-identical waveguides are used for the ring … and Mach-Zehnder input coupler, the Mach-Zehnder arm length difference is to be such that the group delay difference is equal to the round-trip group delay of the ring resonator …, or a multiple … thereof.’ Popovic, figure 13a and paragraphs [0112]-[0113].
Consequently, it would have been obvious to one of ordinary skill in the art to modify Williamson in view of Bandyopadhyay, as applied in the rejection of claims 1-6 and 10-11, to disclose to disclose that the MZI comprises a first branch, a second branch, and a branch length difference between lengths of the first branch and the second branch; Popovic, figure 13a and paragraphs [0112]-[0113]; Williamson, figures 1 and 5, and related figures and text, for example, Selected Text; Bandyopadhyay, figures 1-3, and related figures and text, for example, Selected Text; because the resultant configurations and methods would facilitate designing, fabricating, and deploying coherent optical devices that perform ‘both inference and in situ training;’ by preserving ‘spectral characteristics.’ Popovic, paragraph [0113].
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PETER RADKOWSKI whose telephone number is (571)270-1613. The examiner can normally be reached M-Th 9-5.
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, Thomas Hollweg, can be reached on (571) 270-1739. 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.
/PETER RADKOWSKI/Primary Examiner, Art Unit 2874
Prosecution Insights