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 .
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 12/14/2023, 10/29/2024 and 9/16/2025 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Status of Claims
The present application is being examined under the claims filed on 12/14/2023.
Claims 1-20 are rejected.
Claims 1-20 are pending.
Specification
The specification filed on 12/14/2023 is acceptable for examination purposes.
Drawings
The drawings filed on 12/14/2023 are acceptable for examination purposes.
Claim Objections
Claims 4, 12 and 19 are objected to because of the following informalities:
In claim 4, lines 1-2, “wherein: the temporal-coding-based NN comprises” should read “wherein the temporal-coding-based NN comprises”
In claim 12, lines 1-2, “wherein: the temporal-coding-based NN comprises” should read “wherein the temporal-coding-based NN comprises”
In claim 19, lines 1-2, “wherein: processing the NN components comprises:” should read “wherein processing the NN components comprises:”
In claim 19, line 10, “SSB-NN parameters of the SSB=NN” should read “SSB-NN parameters of the SSB-NN”
Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding Claim 1,
Step 1: Claim 1 is a method claim. Therefore, Claims 1-8 are directed to a process.
Step 2A Prong 1:
processing the NN components to generate scaled NN components (mental process - processing the NN components to generate scaled NN components may be performed manually by a user with the aid of pen and paper by observing/analyzing the NN components and using judgement/evaluation to generate scaled NN components. See MPEP 2106.04(a)(2)(III)(C); Examiner’s note: Paragraph [0073] teaches components are architecture and parameters. Processing the NN parameters and getting the scaled NN parameters are mental processes.)
generating, based at least in part on one or more of the scaled NN components, temporal-coding-based NN components of a temporal-coding-based NN (mental process - generating, based at least in part on one or more of the scaled NN components, temporal-coding-based NN components of a temporal-coding-based NN may be performed manually by a user with the aid of pen and paper by observing/analyzing one or more of the scaled NN components and using judgement/evaluation to generate temporal-coding-based NN components of a temporal-coding-based NN. See MPEP 2106.04(a)(2)(III)(C); Examiner’s note: Paragraph [0073] teaches components are architecture and parameters. Generating temporal-coding-based NN parameters based on the scaled NN parameters is a mental process.)
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
accessing neural network (NN) components of a pre-trained analog-signal-based NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
wherein the temporal-coding-based NN components are dependent on the one or more of the scaled NN components (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
accessing neural network (NN) components of a pre-trained analog-signal-based NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
wherein the temporal-coding-based NN components are dependent on the one or more of the scaled NN components (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
For the reasons above, Claim 1 is rejected as being directed to an abstract idea without significantly more. This rejection applies equally to dependent claims 2-8. The additional limitations of the dependent claims are addressed below.
Regarding Claim 2,
Step 2A Prong 1:
See the rejection of Claim 1 above, which Claim 2 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein the NN components of the pre-trained analog-signal-based NN comprise layers, weights, and activations (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the NN components of the pre-trained analog-signal-based NN comprise layers, weights, and activations (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 3,
Step 2A Prong 1:
wherein processing the NN components comprises: scaling one or more of the weights to generate scaled weights (mathematical concept - wherein processing the NN components comprises: scaling one or more of the weights to generate scaled weights may be performed by mathematical process, using a mathematical algorithm and generating scaled weights. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: specification, paragraph [0080], “the input weights are scaled and then the output weights are scaled (e.g., using line 14 of Algorithm 1 shown in FIG. 6C).”)
Step 2A Prong 2 & Step 2B:
There are no additional elements.
Regarding Claim 4,
Step 2A Prong 1:
See the rejection of Claim 3 above, which Claim 4 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein: the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN); the temporal-coding-based NN components comprise SSB-NN components; and the SSB-NN components comprise SSB-NN parameters of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein: the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN); the temporal-coding-based NN components comprise SSB-NN components; and the SSB-NN components comprise SSB-NN parameters of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 5,
Step 2A Prong 1:
wherein the SSB-NN parameters of the SSB-NN are determined based at least in part on the scaled weights (mathematical concept - wherein the SSB-NN parameters of the SSB-NN are determined based at least in part on the scaled weights may be performed by mathematical process, using a mathematical algorithm and determining the SSB-NN parameters of the SSB-NN using the scaled weights. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: specification, paragraph [0080], “the input weights are scaled and then the output weights are scaled (e.g., using line 14 of Algorithm 1 shown in FIG. 6C).”)
Step 2A Prong 2 & Step 2B:
There are no additional elements.
Regarding Claim 6,
Step 2A Prong 1:
See the rejection of Claim 4 above, which Claim 6 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein the SSB-NN components comprise time intervals for each layer of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the SSB-NN components comprise time intervals for each layer of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 7,
Step 2A Prong 1:
wherein processing the NN components further comprises computing a maximum activation output value of each of the layers (mathematical concept - wherein processing the NN components further comprises computing a maximum activation output value of each of the layers may be performed by mathematical process, using a mathematical algorithm and computing a maximum activation output value of each of the layers. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: specification, paragraph [0080], “block 622, 622A computes the maximum output of this layer for a sample of training data (e.g., using line 15 of Algorithm 1 shown in FIG. 6C).”)
wherein the time intervals for each layer of the SSB-NN are generated based at least in part on the maximum activation output of the respective layer of the SSB-NN (mathematical concept - wherein the time intervals for each layer of the SSB-NN are generated based at least in part on the maximum activation output of the respective layer of the SSB-NN may be performed by mathematical process, using a mathematical algorithm/formula and generating the time intervals for each layer using the maximum activation output of the respective layer of the SSB-NN. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: Fig. 10 and specification, paragraph [0085], “The provided formulas show how to calculate weights of the SNN (Jij), time intervals (tmax) […]”)
Step 2A Prong 2 & Step 2B:
There are no additional elements.
Regarding Claim 8,
Step 2A Prong 1:
See the rejection of Claim 7 above, which Claim 8 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein the SSB-NN comprises a single-spike NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the SSB-NN comprises a single-spike NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 9,
Step 1: Claim 9 is a system claim. Therefore, Claims 9-16 are directed to a machine.
Step 2A Prong 1:
processing the NN components to generate scaled NN components (mental process - processing the NN components to generate scaled NN components may be performed manually by a user with the aid of pen and paper by observing/analyzing the NN components and using judgement/evaluation to generate scaled NN components. See MPEP 2106.04(a)(2)(III)(C).)
generating, based at least in part on one or more of the scaled NN components, temporal-coding-based NN components of a temporal-coding-based NN (mental process - generating, based at least in part on one or more of the scaled NN components, temporal-coding-based NN components of a temporal-coding-based NN may be performed manually by a user with the aid of pen and paper by observing/analyzing one or more of the scaled NN components and using judgement/evaluation to generate temporal-coding-based NN components of a temporal-coding-based NN. See MPEP 2106.04(a)(2)(III)(C).)
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
a processor system electronically coupled to a memory, wherein the processor system is operable to perform processor system operations comprising (merely using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
accessing neural network (NN) components of a pre-trained analog-signal-based NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
wherein the temporal-coding-based NN components are dependent on the one or more of the scaled NN components (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
a processor system electronically coupled to a memory, wherein the processor system is operable to perform processor system operations comprising (merely using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
accessing neural network (NN) components of a pre-trained analog-signal-based NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
wherein the temporal-coding-based NN components are dependent on the one or more of the scaled NN components (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
For the reasons above, Claim 9 is rejected as being directed to an abstract idea without significantly more. This rejection applies equally to dependent claims 10-16. The additional limitations of the dependent claims are addressed below.
Regarding Claim 10,
Step 2A Prong 1:
See the rejection of Claim 9 above, which Claim 10 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein the NN components of the pre-trained analog-signal-based NN comprise layers, weights, and activations (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the NN components of the pre-trained analog-signal-based NN comprise layers, weights, and activations (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 11,
Step 2A Prong 1:
wherein processing the NN components comprises: scaling one or more of the weights to generate scaled weights (mathematical concept - wherein processing the NN components comprises: scaling one or more of the weights to generate scaled weights may be performed by mathematical process, using a mathematical algorithm and generating scaled weights. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: specification, paragraph [0080], “the input weights are scaled and then the output weights are scaled (e.g., using line 14 of Algorithm 1 shown in FIG. 6C).”)
Step 2A Prong 2 & Step 2B:
There are no additional elements.
Regarding Claim 12,
Step 2A Prong 1:
See the rejection of Claim 11 above, which Claim 12 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein: the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN); the temporal-coding-based NN components comprise SSB-NN components; and the SSB-NN components comprise SSB-NN parameters of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein: the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN); the temporal-coding-based NN components comprise SSB-NN components; and the SSB-NN components comprise SSB-NN parameters of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 13,
Step 2A Prong 1:
See the rejection of Claim 12 above, which Claim 13 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein the SSB-NN components further comprise time intervals for each layer of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the SSB-NN components further comprise time intervals for each layer of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 14,
Step 2A Prong 1:
wherein the SSB-NN parameters of the SSB-NN are determined based at least in part on the scaled weights (mathematical concept - wherein the SSB-NN parameters of the SSB-NN are determined based at least in part on the scaled weights may be performed by mathematical process, using a mathematical algorithm and determining the SSB-NN parameters of the SSB-NN using the scaled weights. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: specification, paragraph [0080], “the input weights are scaled and then the output weights are scaled (e.g., using line 14 of Algorithm 1 shown in FIG. 6C).”)
Step 2A Prong 2 & Step 2B:
There are no additional elements.
Regarding Claim 15,
Step 2A Prong 1:
wherein processing the NN components further comprises computing a maximum activation output of each of the layers (mathematical concept - wherein processing the NN components further comprises computing a maximum activation output of each of the layers may be performed by mathematical process, using a mathematical algorithm and computing a maximum activation output of each of the layers. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: specification, paragraph [0080], “block 622, 622A computes the maximum output of this layer for a sample of training data (e.g., using line 15 of Algorithm 1 shown in FIG. 6C).”)
wherein the time intervals for each layer of the SSB-NN are generated based at least in part on the maximum activation output of the respective layer of the SSB-NN (mathematical concept - wherein the time intervals for each layer of the SSB-NN are generated based at least in part on the maximum activation output of the respective layer of the SSB-NN may be performed by mathematical process, using a mathematical algorithm/formula and generating the time intervals for each layer using the maximum activation output of the respective layer of the SSB-NN. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: Fig. 10 and specification, paragraph [0085], “The provided formulas show how to calculate weights of the SNN (Jij), time intervals (tmax) […]”)
Step 2A Prong 2 & Step 2B:
There are no additional elements.
Regarding Claim 16,
Step 2A Prong 1:
See the rejection of Claim 15 above, which Claim 16 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein the SSB-NN comprises a single-spike NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the SSB-NN comprises a single-spike NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 17,
Step 1: Claim 17 is a system claim. Therefore, Claims 17-20 are directed to a machine.
Step 2A Prong 1:
processing the NN components to generate scaled NN components (mental process - processing the NN components to generate scaled NN components may be performed manually by a user with the aid of pen and paper by observing/analyzing the NN components and using judgement/evaluation to generate scaled NN components. See MPEP 2106.04(a)(2)(III)(C).)
generating, based at least in part on one or more of the scaled NN components, temporal-coding-based NN components of a temporal-coding-based NN (mental process - generating, based at least in part on one or more of the scaled NN components, temporal-coding-based NN components of a temporal-coding-based NN may be performed manually by a user with the aid of pen and paper by observing/analyzing one or more of the scaled NN components and using judgement/evaluation to generate temporal-coding-based NN components of a temporal-coding-based NN. See MPEP 2106.04(a)(2)(III)(C).)
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
a computer readable program stored on a computer readable storage medium, wherein the computer readable program, when executed on a processor system, causes the processor system to perform processor system operations comprising (merely using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
accessing neural network (NN) components of a pre-trained analog-signal-based NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
wherein the temporal-coding-based NN components are dependent on the one or more of the scaled NN components (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
a computer readable program stored on a computer readable storage medium, wherein the computer readable program, when executed on a processor system, causes the processor system to perform processor system operations comprising (merely using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
accessing neural network (NN) components of a pre-trained analog-signal-based NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
wherein the temporal-coding-based NN components are dependent on the one or more of the scaled NN components (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
For the reasons above, Claim 17 is rejected as being directed to an abstract idea without significantly more. This rejection applies equally to dependent claims 18-20. The additional limitations of the dependent claims are addressed below.
Regarding Claim 18,
Step 2A Prong 1:
See the rejection of Claim 17 above, which Claim 18 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein the NN components of the pre-trained analog-signal-based NN comprise layers, weights, and activations (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the NN components of the pre-trained analog-signal-based NN comprise layers, weights, and activations (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 19,
Step 2A Prong 1:
wherein: processing the NN components comprises: scaling one or more of the weights to generate scaled weights (mathematical concept – wherein: processing the NN components comprises: scaling one or more of the weights to generate scaled weights may be performed by mathematical process, using a mathematical algorithm and generating scaled weights. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: specification, paragraph [0080], “the input weights are scaled and then the output weights are scaled (e.g., using line 14 of Algorithm 1 shown in FIG. 6C).”)
computing a maximum activation output of each of the layers (mathematical concept - computing a maximum activation output of each of the layers may be performed by mathematical process, using a mathematical algorithm and computing a maximum activation output of each of the layers. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: specification, paragraph [0080], “block 622, 622A computes the maximum output of this layer for a sample of training data (e.g., using line 15 of Algorithm 1 shown in FIG. 6C).”)
the SSB-NN parameters of the SSB-NN are determined based at least in part on the scaled weights (mathematical concept - the SSB-NN parameters of the SSB-NN are determined based at least in part on the scaled weights may be performed by mathematical process, using a mathematical algorithm and determining the SSB-NN parameters of the SSB-NN using the scaled weights. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: specification, paragraph [0080], “the input weights are scaled and then the output weights are scaled (e.g., using line 14 of Algorithm 1 shown in FIG. 6C).”)
the time intervals for each layer of the SSB-NN are generated based at least in part on the maximum activation output of the respective layer of the SSB-NN (mathematical concept - the time intervals for each layer of the SSB-NN are generated based at least in part on the maximum activation output of the respective layer of the SSB-NN may be performed by mathematical process, using a mathematical algorithm/formula and generating the time intervals for each layer using the maximum activation output of the respective layer of the SSB-NN. See MPEP 2106.04(a)(2)(I)(C); Examiner’s note: Fig. 10 and specification, paragraph [0085], “The provided formulas show how to calculate weights of the SNN (Jij), time intervals (tmax) […]”)
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN); the temporal-coding-based NN components comprise SSB-NN components (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
the SSB-NN components comprise: SSB-NN parameters of the SSB=NN; and time intervals for each layer of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN); the temporal-coding-based NN components comprise SSB-NN components (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
the SSB-NN components comprise: SSB-NN parameters of the SSB-NN; and time intervals for each layer of the SSB-NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Regarding Claim 20,
Step 2A Prong 1:
See the rejection of Claim 19 above, which Claim 20 depends on.
Step 2A Prong 2: The judicial exceptions are not integrated into a practical application.
Additional Elements:
wherein the SSB-NN comprises a single-spike NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the SSB-NN comprises a single-spike NN (merely reciting the words "apply it" (or an equivalent) with the judicial exception. See MPEP 2106.05(f).)
Claim Rejections - 35 USC § 102
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 (i.e., changing from AIA to pre-AIA ) 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.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-5, 9-12, 17 and 18 are rejected under 35 U.S.C. 102 as being anticipated by Rueckauer et al (“Conversion of Continuous-Valued Deep Networks to Efficient Event-Driven Networks for Image Classification”), hereinafter Rueckauer.
Regarding Claim 1,
Rueckauer teaches:
“A computer-implemented method comprising:” (preamble)
“accessing neural network (NN) components of a pre-trained analog-signal-based NN” (Rueckauer, Section 1, “A more straightforward approach is to take the parameters of a pre-trained ANN and to map them to an equivalent-accurate SNN.”; Examiner’s note: accessing neural network (NN) components of a pre-trained analog-signal-based NN (i.e. taking the parameters of a pre-trained ANN and mapping them to an equivalent-accurate SNN) is taught.)
“processing the NN components to generate scaled NN components” (Rueckauer, Section 2.2.2, “The data-based weight normalization mechanism is based on the linearity of the ReLU unit used for ANNs. It can simply be extended to biases by linearly rescaling all weights and biases such that the ANN activation a [as computed in Equation (1)] is smaller than 1 for all training examples.”; Examiner’s note: processing the NN components (i.e. rescaling weights and biases) to generate scaled NN components (i.e. rescaled weights and biases) is taught.)
“generating, based at least in part on one or more of the scaled NN components, temporal-coding-based NN components of a temporal-coding-based NN” (Rueckauer, Equation 2 and Section 2.1, “The basic principle of converting ANNs into SNNs is that firing rates of spiking neurons should match the graded activations of analog neurons […] Each SNN neuron has a membrane potential Vli(t), which integrates its input current at every time step:
PNG
media_image1.png
53
269
media_image1.png
Greyscale
where Vthr is the threshold and
PNG
media_image2.png
20
22
media_image2.png
Greyscale
is a step function indicating the occurrence of a spike at time t […]”; Examiner’s note: generating, based at least in part on one or more of the scaled NN components (i.e. weights and biases), temporal-coding-based NN components (i.e. spiking neurons, spike time, weight and the threshold) of a temporal-coding-based NN (i.e. SNN) is taught.)
“wherein the temporal-coding-based NN components are dependent on the one or more of the scaled NN components” (Rueckauer, Equation 2 and Section 2.1, “The basic principle of converting ANNs into SNNs is that firing rates of spiking neurons should match the graded activations of analog neurons […] Each SNN neuron has a membrane potential Vli(t), which integrates its input current at every time step:
PNG
media_image1.png
53
269
media_image1.png
Greyscale
where Vthr is the threshold and
PNG
media_image2.png
20
22
media_image2.png
Greyscale
is a step function indicating the occurrence of a spike at time t […]”; Rueckauer, Section 2.2.2, “The data-based weight normalization mechanism is based on the linearity of the ReLU unit used for ANNs. It can simply be extended to biases by linearly rescaling all weights and biases such that the ANN activation a [as computed in Equation (1)] is smaller than 1 for all training examples.”; Examiner’s note: wherein the temporal-coding-based NN components (i.e. spiking neurons, spike time, weight and threshold) are dependent on the one or more of the scaled NN components (i.e. rescaled weights and biases) is taught.)
Regarding Claim 2,
Rueckauer teaches:
“The computer-implemented method of claim 1,” (preamble)
“wherein the NN components of the pre-trained analog-signal-based NN comprise layers, weights, and activations” (Rueckauer, Section 2.1, “For a network with L layers let Wl, l ∈ {1,...,L} denote the weight matrix connecting units in layer l − 1 to layer l, with biases bl. The number of units in each layer is Ml. The ReLU activation of the continuous-valued neuron i in layer l is computed […]”; Examiner’s note: wherein the NN components of the pre-trained analog-signal-based NN (ANN) comprise layers (i.e. L layers), weights (i.e. the weight matrix connecting units), and activations (i.e. ReLU activation) is taught.)
Regarding Claim 3,
Rueckauer teaches:
“The computer-implemented method of claim 2,” (preamble)
“wherein processing the NN components comprises: scaling one or more of the weights to generate scaled weights” (Rueckauer, Section 2.2.2, “Denoting the maximum ReLU activation in layer l as λl = max[al], then weights Wl and biases bl are normalized to Wl → Wlλl−1 / λl and bl → bl / λl.”; Examiner’s note: wherein processing the NN components comprises: scaling one or more of the weights (i.e. weights normalization) to generate scaled weights (i.e. normalized weights) is taught.)
Regarding Claim 4,
Rueckauer teaches:
“The computer-implemented method of claim 3,” (preamble)
“wherein: the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN); the temporal-coding-based NN components comprise SSB-NN components; and the SSB-NN components comprise SSB-NN parameters of the SSB-NN” (Rueckauer, Equation 2 and Section 2.1, “The basic principle of converting ANNs into SNNs is that firing rates of spiking neurons should match the graded activations of analog neurons […] Each SNN neuron has a membrane potential Vli(t), which integrates its input current at every time step:
PNG
media_image1.png
53
269
media_image1.png
Greyscale
where Vthr is the threshold and
PNG
media_image2.png
20
22
media_image2.png
Greyscale
is a step function indicating the occurrence of a spike at time t […]”; Examiner’s note: wherein: the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN) (i.e. SNN); the temporal-coding-based NN components comprise SSB-NN components (i.e. spiking neurons, spike time, weight and the threshold); and the SSB-NN components comprise SSB-NN parameters (i.e. membrance potential, the threshold and time t) of the SSB-NN (i.e. SNN) is taught.)
Regarding Claim 5,
Rueckauer teaches:
“The computer-implemented method of claim 4,” (preamble)
“wherein the SSB-NN parameters of the SSB-NN are determined based at least in part on the scaled weights” (Rueckauer, Section 2.2.2, “Weight normalization is introduced by Diehl et al. (2015) as a means to avoid approximation errors due to too low or too high firing. This work showed significant improvement of the performance of converted SNNs by using a data-based weight normalization mechanism. We extend this method to the case of neurons with biases […]”; Examiner’s note: wherein the SSB-NN parameters (i.e. weights and neurons with biases) of the SSB-NN (i.e. SNNs) are determined based at least in part on the scaled weights (i.e. weight normalization mechanism) is taught.)
Regarding Claim 9,
Rueckauer teaches:
“A computer system comprising a processor system electronically coupled to a memory, wherein the processor system is operable to perform processor system operations comprising:” (preamble)
“accessing neural network (NN) components of a pre-trained analog-signal-based NN” (Rueckauer – see supra claim 1)
“processing the NN components to generate scaled NN components” (Rueckauer – see supra claim 1)
“generating, based at least in part on one or more of the scaled NN components, temporal-coding-based NN components of a temporal-coding-based NN” (Rueckauer – see supra claim 1)
“wherein the temporal-coding-based NN components are dependent on the one or more of the scaled NN components” (Rueckauer – see supra claim 1)
Regarding Claim 10,
Rueckauer teaches:
“The computer system of claim 9,” (preamble)
“wherein the NN components of the pre-trained analog-signal-based NN comprise layers, weights, and activations” (Rueckauer – see supra claim 2)
Regarding Claim 11,
Rueckauer teaches:
“The computer system of claim 10,” (preamble)
“wherein processing the NN components comprises: scaling one or more of the weights to generate scaled weights” (Rueckauer – see supra claim 3)
Regarding Claim 12,
Rueckauer teaches:
“The computer system of claim 11,” (preamble)
“wherein: the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN); the temporal-coding-based NN components comprise SSB-NN components; and the SSB-NN components comprise SSB-NN parameters of the SSB-NN” (Rueckauer – see supra claim 4)
Regarding Claim 17,
Rueckauer teaches:
“A computer program product comprising a computer readable program stored on a computer readable storage medium, wherein the computer readable program, when executed on a processor system, causes the processor system to perform processor system operations comprising:” (preamble)
“accessing neural network (NN) components of a pre-trained analog-signal-based NN” (Rueckauer – see supra claim 1)
“processing the NN components to generate scaled NN components” (Rueckauer – see supra claim 1)
“generating, based at least in part on one or more of the scaled NN components, temporal-coding-based NN components of a temporal-coding-based NN” (Rueckauer – see supra claim 1)
“wherein the temporal-coding-based NN components are dependent on the one or more of the scaled NN components” (Rueckauer – see supra claim 1)
Regarding Claim 18,
Rueckauer teaches:
“The computer program product of claim 17,” (preamble)
“wherein the NN components of the pre-trained analog-signal-based NN comprise layers, weights, and activations” (Rueckauer – see supra claim 2)
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 6-8, 13-16, 19 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Rueckauer, in view of Park et al. (“T2FSNN: Deep Spiking Neural Networks with Time-to-first-spike Coding”) (hereinafter Park).
Regarding Claim 6,
Rueckauer teaches:
“The computer-implemented method of claim 4,” (preamble)
Rueckauer does not explicitly teach:
“wherein the SSB-NN components comprise time intervals for each layer of the SSB-NN”
Park teaches “T2FSNN: Deep Spiking Neural Networks with Time-to-first-spike Coding (title)” comprising:
“wherein the SSB-NN components comprise time intervals for each layer of the SSB-NN” (Park, Fig. 3 and Section III, “In each layer of this baseline pipeline (Fig. 3-(a), T2FSNN), the integration and fire phases in a layer are executed sequentially. The integration phase of layer l is executed simultaneously with the fire phase of the previous layer l - 1 from (L - 1)T to LT time step […] we propose a technique called early firing, which is a method of starting the firing phase before the integration is complete at each layer. By using this method, we can overlap the integration and fire phase, which results in reducing inference latency […]“; Examiner’s note: wherein the SSB-NN (i.e. T2FSNN) components comprise time intervals for each layer (i.e. sequential phase interval and early firing interval) of the SSB-NN is taught.
PNG
media_image3.png
295
343
media_image3.png
Greyscale
)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to combine the spiking neural network (SNN) in Rueckauer, and the T2FSNN: Deep Spiking Neural Networks with Time-to-first-spike Coding as taught in Park. Rueckauer teaches converting the analog-signal-based NN (ANN) to the spiking neural network (SNN). Park teaches the SSB-NN components comprising time intervals for each layer of the SSB-NN. One of ordinary skill would have motivation to combine Rueckauer and Park to “efficiently implement[] TTFS coding in deep SNNs […] further increase[] accuracy and reduce[] inference latency with the feature of TTFS coding” (Park, Section III).
Regarding Claim 7,
The combination of Rueckauer and Park teaches:
“The computer-implemented method of claim 6,” (preamble)
“wherein processing the NN components further comprises computing a maximum activation output value of each of the layers” (Rueckauer, Section 2.2.2, “[…] the normalization factor λl was set to the maximum ANN activation within a layer, where the activations are computed using a large subset of the training data.”; Examiner’s note: wherein processing the NN components further comprises computing a maximum activation (i.e. the maximum ANN activation) output value of each of the layers (i.e. a large subset of the training data) is taught.)
wherein the time intervals for each layer of the SSB-NN are generated based at least in part on the respective layer of the SSB-NN (Park – see supra claim 6.)
the maximum activation output of the respective layer of the SSB-NN (Rueckauer, Section 2.2.2, “[…] the normalization factor λl was set to the maximum ANN activation within a layer, where the activations are computed using a large subset of the training data.”; Examiner’s note: the maximum activation (i.e. the maximum ANN activation) output of the respective layer (i.e. a large subset of the training data) of the SSB-NN (i.e. T2FSNN) is taught.)
The reasons of obviousness have been noted in the rejection of Claim 6 above and applicable herein.
Regarding Claim 8,
The combination of Rueckauer and Park teaches:
“The computer-implemented method of claim 7,” (preamble)
“wherein the SSB-NN comprises a single-spike NN” (Park, Section 4, “[…] an approach where the precise spike time is used to train a network to classify MNIST digits with a single spike per neuron.”; Examiner’s note: wherein the SSB-NN (i.e. T2FSNN) comprises a single-spike NN (i.e. a single spike per neuron) is taught.)
The reasons of obviousness have been noted in the rejection of Claim 6 above and applicable herein.
Regarding Claim 13,
The combination of Rueckauer and Park teaches:
“The computer system of claim 12,” (preamble)
“wherein the SSB-NN components further comprise time intervals for each layer of the SSB-NN” (Park – see supra claim 6)
The reasons of obviousness have been noted in the rejection of Claim 6 above and applicable herein.
Regarding Claim 14,
The combination of Rueckauer and Park teaches:
“The computer system of claim 13,” (preamble)
“wherein the SSB-NN parameters of the SSB-NN are determined based at least in part on the scaled weights” (Rueckauer – see supra claim 5)
The reasons of obviousness have been noted in the rejection of Claim 6 above and applicable herein.
Regarding Claim 15,
The combination of Rueckauer and Park teaches:
“The computer system of claim 13,” (preamble)
“wherein processing the NN components further comprises computing a maximum activation output of each of the layers” (Rueckauer – see supra claim 7)
wherein the time intervals for each layer of the SSB-NN are generated based at least in part on the respective layer of the SSB-NN (Park – see supra claim 6.)
the maximum activation output of the respective layer of the SSB-NN (Rueckauer – see supra claim 7)
The reasons of obviousness have been noted in the rejection of Claim 6 above and applicable herein.
Regarding Claim 16,
The combination of Rueckauer and Park teaches:
“The computer system of claim 15,” (preamble)
“wherein the SSB-NN comprises a single-spike NN” (Park – see supra claim 8)
The reasons of obviousness have been noted in the rejection of Claim 6 above and applicable herein.
Regarding Claim 19,
The combination of Rueckauer and Park teaches:
“The computer program product of claim 18,” (preamble)
“wherein: processing the NN components comprises: scaling one or more of the weights to generate scaled weights” (Rueckauer – see supra claim 3)
“computing a maximum activation output of each of the layers” (Rueckauer – see supra claim 7)
“the temporal-coding-based NN comprises a spike-signal-based NN (SSB-NN); the temporal-coding-based NN components comprise SSB-NN components; the SSB-NN components comprise: SSB-NN parameters of the SSB=NN” (Rueckauer – see supra claim 4)
“time intervals for each layer of the SSB-NN” (Park – see supra claim 6)
“the SSB-NN parameters of the SSB-NN are determined based at least in part on the scaled weights” (Rueckauer – see supra claim 5)
the time intervals for each layer of the SSB-NN are generated based at least in part on the respective layer of the SSB-NN (Park – see supra claim 6.)
the maximum activation output of the respective layer of the SSB-NN (Rueckauer – see supra claim 7)
The reasons of obviousness have been noted in the rejection of Claim 6 above and applicable herein.
Regarding Claim 20,
The combination of Rueckauer and Park teaches:
“The computer program product of claim 19,” (preamble)
“wherein the SSB-NN comprises a single-spike NN” (Park – see supra claim 8)
The reasons of obviousness have been noted in the rejection of Claim 6 above and applicable herein.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to YONG D RHO whose telephone number is (571)270-0194. The examiner can normally be reached 8am-5pm.
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, Viker Lamardo can be reached at 5712705871. 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.
/YONG DOO RHO/Examiner, Art Unit 2147
/VIKER A LAMARDO/Supervisory Patent Examiner, Art Unit 2147
Prosecution Insights