DETAILED ACTION
This action is in response to the Applicant Response filed 17 October 2025 for application 17/143,796 filed 07 January 2021.
Claim(s) 34-53 is/are new.
Claim(s) 1-33 is/are cancelled.
Claim(s) 34-53 is/are pending.
Claim(s) 34-53 is/are rejected.
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
Applicant's arguments regarding the objections to the claims have been fully considered and, in light of the amendments to the claims, are persuasive. However, in light of the amendments to the claims, new claim objections have arisen, as noted below.
Applicant's arguments regarding the 35 U.S.C. 112(b) rejection(s) of claim(s) 21-24 have been fully considered and, in light of the amendments to the claims, are persuasive. The 35 U.S.C. 112(b) rejection(s) of claim(s) 21-24 has/have been withdrawn.
Applicant’s arguments regarding the 35 U.S.C. 101 rejection of the claims are based on the newly amended subject matter. All arguments are addressed in the 35 U.S.C. 101 rejection of the claims below.
Applicant’s arguments regarding the 35 U.S.C. 102 and/or 35 U.S.C. 103 rejections of the claims are based on the newly amended subject matter. All arguments are addressed in the 35 U.S.C. 102 and/or 35 U.S.C. 103 rejections of the claims below.
Claim Objections
Claim(s) 49-53 is/are objected to because of the following informalities:
Claim 49, line 2, the processor should read “the at least one processor”
Claim 49, line 3, the processor should read “the at least one processor”
Claim 49, line 9, the network should read “the neural network”
Claim 53, line 2, the processors should read “the one or more processors”
Claims 50-52 are objected to due to their dependence, either directly or indirectly, on claims 49, 53
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.
Claim(s) 34-53 is/are rejected under 35 U.S.C. 101, because the claim(s) is/are directed to an abstract idea, and because the claim elements, whether considered individually or in combination, do not amount to significantly more than the abstract idea, see Alice Corporation Pty. Ltd. V. CLS Bank International et al., 573 US 208 (2014).
Regarding claim 34, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 34 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices.
The limitation of creating, for each device, a corresponding neuron in the neural network, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of embedding, within each neuron, the one or more device-behavior equations associated with a corresponding device as an activation function of that neuron, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of connecting neurons according to relationships among the plurality of devices so that: inputs to each neuron correspond to outputs from nodes representing related devices, and outputs of each neuron correspond to transformed variables generated by evaluating its activation function, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – computer-implemented, system of linked devices, computing system, plurality of devices. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
The claim recites additional element(s) – neural network, one or more device-behavior equations, activation function, cost function. The additional element(s) is/are recited at a high-level of generality such that it amounts to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h)).
The claim recites receiving, by a computing system, a linked network definition specifying: a plurality of devices, for each device, one or more device-behavior equations that model physical behavior of the device, which is simply acquiring data recited at a high level of generality. This is nothing more than insignificant extra-solution activity (MPEP 2106.05(g)).
The claim recites executing the neural network to compute values of variables within the neurons by evaluating, within each neuron, the activation function which is simply applying the model recited at a high level of generality and amounts to the recitation of the words “apply it” (or an equivalent) or amounts to no more than mere instructions to implement an abstract idea or other exception on a computer (MPEP 2106.05(f)).
The claim recites computing a derivative of the neural network to minimize a cost function that measures a difference between network output and real-world data which is simply generic training to perform the abstract idea of model construction and amounts to mere instructions to apply the exception (MPEP 2106.05(f)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
computer-implemented, system of linked devices, computing system, plurality of devices amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
applying the model and generic training to perform the abstract idea amount(s) to no more than mere instructions to apply the exception (MPEP 2106.05(f))
acquiring data amount(s) to no more than insignificant extra-solution activity (MPEP 2106.05(g)), wherein the insignificant extra-solution activity is the well-understood routine and conventional activit(y/ies) of receiving or transmitting data over a network and/or storing and retrieving information in memory (MPEP 2016.05(d))
neural network, one or more device-behavior equations, activation function amount(s) to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 35, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 35 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 35 carries out the method of claim 34 but for the recitation of additional element(s) of wherein the one or more device-behavior equations embedded within at least some neurons are differentiable with respect to their inputs and internal parameters.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the device-behavior equations and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the device-behavior equations do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 36, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 36 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 36 carries out the method of claim 34 but for the recitation of additional element(s) of wherein at least two neurons comprise different activation functions that implement unrelated device-behavior equations.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the neurons and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the neurons do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 37, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 37 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 37 carries out the method of claim 34 but for the recitation of additional element(s) of wherein at least one neuron has an activation function comprising a plurality of equations representing distinct physical properties of the corresponding device.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the neurons and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the neurons do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 38, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 38 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices.
The limitation of creating, for each neuron, one or more inputs representing internal properties of the corresponding device, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated
into a practical application. The claim does not recite any additional elements which integrate the
abstract idea into a practical application and, therefore, does not impose any meaningful limits on
practicing the abstract idea. Therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to
significantly more than the judicial exception. As discussed above with respect to the integration of the
abstract idea into a practical application, the claim does not recite any additional elements which
provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 39, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 39 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 39 carries out the method of claim 34 but for the recitation of additional element(s) of wherein each edge between neurons carries a variable selected from a group consisting of electrical variables including voltage and current, mechanical variables including torque and angular velocity, and fluid variables including specific enthalpy, mass flow rate, and pressure.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the edges and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the edges do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 40, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 40 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 40 carries out the method of claim 34 but for the recitation of additional element(s) of wherein a first neuron has a first number of upstream connections and a different number of downstream connections corresponding to physical relationships among the one or more devices represented by the neurons.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the neurons and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the neurons do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 41, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 41 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 41 carries out the method of claim 34 but for the recitation of additional element(s) of wherein executing the neural network comprises simulating a physical system over successive timesteps such that neuron outputs correspond to evolving states of the one or more devices represented by the neurons.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – physical system. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
The claim recites wherein executing the neural network comprises simulating a physical system over successive timesteps such that neuron outputs correspond to evolving states of the one or more devices represented by the neurons which is simply applying the model recited at a high level of generality and amounts to the recitation of the words “apply it” (or an equivalent) or amounts to no more than mere instructions to implement an abstract idea or other exception on a computer (MPEP 2106.05(f)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
physical system amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
applying the model amount(s) to no more than mere instructions to apply the exception (MPEP 2106.05(f))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 42, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 42 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 42 carries out the method of claim 34 but for the recitation of additional element(s) of wherein the cost function measures a distance between neural-network-generated output variables and measured data from an actual physical system modeled by the neural network.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the cost function and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the cost function do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 43, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 43 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 43 carries out the method of claim 34 but for the recitation of additional element(s) of wherein computing the derivative of the neural network applies only to a subset of neuron inputs including permanent property inputs or temporary variable inputs that represent system states.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the neurons and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the neurons do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 44, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 44 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 44 carries out the method of claim 34 but for the recitation of additional element(s) of wherein the neural network comprises neurons corresponding to at least two of a sensor, a pump, a boiler, a valve, a relay, a motor, or a heating coil.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the neural network and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the neural network do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 45, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 45 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 45 carries out the method of claim 34 but for the recitation of additional element(s) of wherein at least some of the activation functions are non-trainable and compute physical equations directly using known parameters of the corresponding devices.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the activation functions and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the activation functions do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 46, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 46 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 46 carries out the method of claim 34 but for the recitation of additional element(s) of wherein computing the derivative of the neural network is performed by automatic differentiation applied to the activation functions.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the activation functions and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the activation functions do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 47, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 47 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 47 carries out the method of claim 34 but for the recitation of additional element(s) of wherein each weight assigned to an edge corresponds to a real physical variable value transferred between modeled devices.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the weights and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the weights do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 48, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 48 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-implemented method for constructing a neural network that models a system of linked devices. The Step 2A Prong One Analysis for claim 34 is applicable here since claim 48 carries out the method of claim 34 but for the recitation of additional element(s) of wherein embedding device-behavior equations as activation functions reduces or eliminates a need for training using large data sets.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the device-behavior equations and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the device-behavior equations do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 49, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 49 is directed to a system with a processor, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) system for constructing a neural network that models a linked system of devices.
The limitation of create a neuron corresponding to each device, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of embed one or more device-behavior equations for each device within the neuron as an activation function, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of connect the neurons based on relationships among the devices so that input and output variables are exchanged as weighted edges, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – system, linked system of devices, at least one processor, memory, instructions. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
The claim recites additional element(s) – neural network, device-behavior equations, activation function, cost function. The additional element(s) is/are recited at a high-level of generality such that it amounts to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h)).
The claim recites execute the network to simulate device behavior ... which is simply applying the model recited at a high level of generality and amounts to the recitation of the words “apply it” (or an equivalent) or amounts to no more than mere instructions to implement an abstract idea or other exception on a computer (MPEP 2106.05(f)).
The claim recites execute the network to ... compute derivatives of the network with respect to selected inputs to minimize a cost function which is simply generic training to perform the abstract idea of model construction and amounts to mere instructions to apply the exception (MPEP 2106.05(f)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
system, linked system of devices, at least one processor, memory, instructions amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
applying the model and generic training to perform the abstract idea amount(s) to no more than mere instructions to apply the exception (MPEP 2106.05(f))
neural network, device-behavior equations, activation function, cost function amount(s) to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 50, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 50 is directed to a system with a processor, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) system for constructing a neural network that models a linked system of devices. The Step 2A Prong One Analysis for claim 49 is applicable here since claim 50 carries out the system of claim 49 but for the recitation of additional element(s) of wherein the neurons comprise permanent inputs representing device properties and temporary inputs representing state variables and cost function optimization selectively differentiates with respect to one or both input types.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the neurons and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the neurons do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 51, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 51 is directed to a system with a processor, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) system for constructing a neural network that models a linked system of devices. The Step 2A Prong One Analysis for claim 49 is applicable here since claim 51 carries out the system of claim 49 but for the recitation of additional element(s) of wherein each neuron corresponds to a physical device within a real-world system and outputs of the neurons represent predicted physical behaviors of the devices.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites wherein each neuron corresponds to a physical device within a real-world system and outputs of the neurons represent predicted physical behaviors of the devices which is simply additional information regarding the neurons, and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)).
The claim recites additional element(s) – physical device, real-world system. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
physical device, real-world system amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
additional information regarding the neurons do(es) not apply the exception in a meaningful way (MPEP 2106.05(e))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 52, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 52 is directed to a system with a processor, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) system for constructing a neural network that models a linked system of devices. The Step 2A Prong One Analysis for claim 49 is applicable here since claim 52 carries out the system of claim 49 but for the recitation of additional element(s) of wherein at least two neurons have activation functions representing unrelated physical processes.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application. In particular, the claim recites additional information regarding the neurons and the element(s) do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of additional information regarding the neurons do(es) not apply the exception in a meaningful way (MPEP 2106.05(e)). Not applying the exception in a meaningful way does not provide an inventive concept, and, therefore, the claim is not patent eligible.
Regarding claim 53, the claim is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 Analysis: Claim 53 is directed to a computer-readable medium, which is directed to an article of manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The claim recites a(n) computer-readable medium.
The limitation of creating, for each device a corresponding neuron in a neural network, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of embedding, within each neuron, the device-behavior equations associated with the corresponding device as an activation function of that neuron, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
The limitation of connecting neurons according to relationships among the devices so that: inputs to each neuron correspond to outputs from nodes representing related devices, and outputs of each neuron correspond to transformed variables generated by evaluating its activation function, as drafted, is a process that, under its broadest reasonable interpretation, covers a mental process. The limitation is directed to observation, evaluation, judgment and opinion and is a process capable of being performed by a human mentally or using pen and paper.
If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind, then it falls within the "Mental Processes" grouping. Accordingly, the claim recites an abstract idea.
Step 2A Prong Two Analysis: With respect to the abstract idea, the judicial exception is not integrated into a practical application.
The claim recites additional element(s) – computer-readable medium, instructions, one or more processors, computing system, linked network, plurality of devices. The additional element(s) is/are recited at a high-level of generality (i.e., as generic computer components performing generic computer functions of executing instructions on the computers) such that it amounts to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b)).
The claim recites additional element(s) – neural network, one or more device-behavior equations, activation function, cost function. The additional element(s) is/are recited at a high-level of generality such that it amounts to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h)).
The claim recites receiving, by a computing system, a linked network definition specifying: a plurality of devices, for each device, one or more device-behavior equations that model physical behavior of the device, which is simply acquiring data recited at a high level of generality. This is nothing more than insignificant extra-solution activity (MPEP 2106.05(g)).
The claim recites executing the neural network to compute values of variables by evaluating, within each neuron, the embedded device-behavior equations which is simply applying the model recited at a high level of generality and amounts to the recitation of the words “apply it” (or an equivalent) or amounts to no more than mere instructions to implement an abstract idea or other exception on a computer (MPEP 2106.05(f)).
The claim recites computing a derivative of the neural network to minimize a cost function that measures a difference between network output and real-world data which is simply generic training to perform the abstract idea of model construction and amounts to mere instructions to apply the exception (MPEP 2106.05(f)).
Accordingly, the additional element(s) do(es) not integrate the abstract idea into a practical application because the additional element(s) do(es) not impose any meaningful limits on practicing the abstract idea, and, therefore, the claim is directed to an abstract idea.
Step 2B Analysis: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to the integration of the abstract idea into a practical application, the additional element(s) of:
computer-implemented, system of linked devices, computing system, plurality of devices amount(s) to no more than mere instructions to apply the exception using generic computer components (MPEP 2106.05(b))
applying the model and generic training to perform the abstract idea amount(s) to no more than mere instructions to apply the exception (MPEP 2106.05(f))
acquiring data amount(s) to no more than insignificant extra-solution activity (MPEP 2106.05(g)), wherein the insignificant extra-solution activity is the well-understood routine and conventional activit(y/ies) of receiving or transmitting data over a network and/or storing and retrieving information in memory (MPEP 2016.05(d))
neural network, one or more device-behavior equations, activation function amount(s) to no more than indicating a field of use or technological environment in which to apply the judicial exception (MPEP 2106.05(h))
The additional element(s) do(es) not provide an inventive concept, and, therefore, the claim is not patent eligible.
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.
The factual inquiries 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.
Claim(s) 34-53 is/are rejected under 35 U.S.C. 103 as being unpatentable over Deshpande et al. (Model-Driven Data Acquisition in Sensor Networks, hereinafter referred to as “Deshpande”) in view of Tulone et al. (PAQ: Time Series Forecasting for Approximate Query Answering in Sensor Networks, hereinafter referred to as Tulone”).
Regarding claim 34 (New), Deshpande teaches a computer-implemented (Deshpande, section 2 – teaches operating code [requiring memory] on a query processor) method for constructing a neural network that models a system of linked devices (Deshpande, section 2 – teaches modeling the behavior a sensor network with sensor nodes [neurons]), the method comprising:
receiving, by a computing system (Deshpande, section 2 – teaches operating code [requiring memory] on a query processor), a linked network definition (Deshpande, section 2 – teaches modeling the behavior a sensor network with sensor nodes [neurons]) specifying:
a plurality of devices (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes),
for each device, one or more device-behavior equations that model physical behavior of the device (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes), and
creating, for each device, a corresponding neuron in the neural network (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes);
embedding, within each neuron, the one or more device-behavior equations associated with a corresponding device as an activation function of that neuron (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes);
connecting neurons according to relationships among the plurality of devices (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes) so that:
inputs to each neuron correspond to outputs from nodes representing related devices (Deshpande, section 4.1 – teaches edges as weighted probabilities between nodes), and
outputs of each neuron correspond to transformed variables generated by evaluating its activation function (Deshpande, section 4.1 – teaches edges as weighted probabilities between nodes);
executing the neural network to compute values of variables within the neurons by evaluating, within each neuron, the activation function (Deshpande, sections 2-3 – teaches running the network model to model the behavior of the devices); and
computing a derivative of the neural network (Deshpande, sections 2-3 – teaches running the network model to model the behavior of the devices) to minimize a cost function that measures a difference between network output and real-world data (Deshpande, section 4.3 – teaches minimizing a cost function).
While Deshpande teaches a network of sensor nodes and modeling the nodes based on an activation function, Deshpande does not explicitly teach an activation function/equation for each device/neuron.
Tulone teaches
receiving, by a computing system, a linked network definition specifying:
a plurality of devices (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device),
for each device, one or more device-behavior equations that model physical behavior of the device (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device), and
embedding, within each neuron, the one or more device-behavior equations associated with a corresponding device as an activation function of that neuron (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device);
connecting neurons according to relationships among the plurality of devices so that:
inputs to each neuron correspond to outputs from nodes representing related devices, and
outputs of each neuron correspond to transformed variables generated by evaluating its activation function (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device);
executing the neural network to compute values of variables within the neurons by evaluating, within each neuron, the activation function (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to modify Deshpande with the teachings of Tulone in order to improve efficiency in the field of heterogeneous sensor neural networks (Tulone, section 4.1 – “... For these reasons, it is more efficient for the sensor to compute m univariate AR models, although the multivariate model provides additional predictive power as it is able to capture correlations between measurements...”).
Regarding claim 35 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Tulone further teaches wherein the one or more device-behavior equations embedded within at least some neurons are differentiable with respect to their inputs and internal parameters (Tulone, section 4.2 – teaches the sensor functions are autoregressive models [which are differentiable]).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Deshpande and Tulone in order to use differentiable functions to the equations to improve efficiency and capture data correlations (Tulone, section 4.1).
Regarding claim 36 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Tulone further teaches wherein at least two neurons comprise different activation functions that implement unrelated device-behavior equations (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Deshpande and Tulone in order to use attribute inputs to the equations to improve efficiency and capture data correlations (Tulone, section 4.1).
Regarding claim 37 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Tulone further teaches wherein at least one neuron has an activation function comprising a plurality of equations representing distinct physical properties of the corresponding device (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Deshpande and Tulone in order to use attribute inputs to the equations to improve efficiency and capture data correlations (Tulone, section 4.1).
Regarding claim 38 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Tulone further teaches creating, for each neuron, one or more inputs representing internal properties of the corresponding device (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Deshpande and Tulone in order to use multiple properties with multiple values to improve efficiency and capture data correlations (Tulone, section 4.1).
Regarding claim 39 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Deshpande teaches wherein each edge between neurons carries a variable (Deshpande, section 4.1 – teaches transmitting data between sensors [neurons]) selected from a group consisting of electrical variables including voltage and current, mechanical variables including torque and angular velocity, and fluid variables including specific enthalpy, mass flow rate, and pressure (Deshpande, section 2 – teaches variables including voltage).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 34 above.
Regarding claim 40 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Deshpande further teaches wherein a first neuron has a first number of upstream connections and a different number of downstream connections corresponding to physical relationships among the one or more devices represented by the neurons (Deshpande, section 4 - teaches edges connecting nodes in the network and selecting a path [A path means upstream and downstream nodes]).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 34 above.
Regarding claim 41 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Deshpande further teaches wherein executing the neural network comprises simulating a physical system (Deshpande, section 2- teaches a physical system) over successive timesteps such that neuron outputs correspond to evolving states of the one or more devices represented by the neurons (Deshpande, section 3.2 – teaches generating and outputting data over time; see also Deshpande, section 2).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 34 above.
Regarding claim 42 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Deshpande further teaches wherein the cost function measures a distance between neural-network-generated output variables and measured data from an actual physical system modeled by the neural network (Deshpande, section 4 – teaches optimizing the cost based on the difference of predictions and real world measurements).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 34 above.
Regarding claim 43 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Deshpande further teaches wherein computing the derivative of the neural network applies only to a subset of neuron inputs including permanent property inputs or temporary variable inputs that represent system states (Deshpande, section 4.3 – teaches a cost function based on attributes and data transfers of a subset of the sensor network).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 34 above.
Regarding claim 44 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Deshpande further teaches wherein the neural network comprises neurons corresponding to at least two of a sensor, a pump, a boiler, a valve, a relay, a motor, or a heating coil (Deshpande, sections 2-3 – teaches a network of sensors [at least 2 sensors]).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 34 above.
Regarding claim 45 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Tulone further teaches wherein at least some of the activation functions are non-trainable and compute physical equations directly using known parameters of the corresponding devices (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Deshpande and Tulone in order to use multiple properties with multiple values to improve efficiency and capture data correlations (Tulone, section 4.1).
Regarding claim 46 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Tulone further teaches wherein computing the derivative of the neural network is performed by automatic differentiation applied to the activation functions (Tulone, section 4.2 – teaches the sensor functions are autoregressive models [which are differentiable]).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Deshpande and Tulone in order to use differentiable functions to the equations to improve efficiency and capture data correlations (Tulone, section 4.1).
Regarding claim 47 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Deshpande further teaches wherein each weight assigned to an edge corresponds to a real physical variable value transferred between modeled devices (Deshpande, section 4.1 – teaches edge probabilities of physical value transfer between sensors [neurons]).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 34 above.
Regarding claim 48 (New), Deshpande in view of Tulone teaches all of the limitations of the method of claim 34 as noted above. Deshpande further teaches wherein embedding device-behavior equations as activation functions reduces or eliminates a need for training using large data sets (Deshpande, section 3.2 – teaches reducing the amount of data required).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 34 above.
Regarding claim 49 (New), Deshpande teaches a system for constructing a neural network that models a linked system of devices (Deshpande, section 2 – teaches modeling the behavior a sensor network with sensor nodes [neurons]), comprising at least one processor and a memory coupled to the processor storing executable instructions (Deshpande, section 2 – teaches operating code [requiring memory] on a query processor) that, when executed, cause the processor to:
create a neuron corresponding to each device (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes);
embed one or more device-behavior equations for each device within the neuron as an activation function (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes);
connect the neurons based on relationships among the devices (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes) so that input and output variables are exchanged as weighted edges (Deshpande, section 4.1 – teaches edges as weighted probabilities between nodes); and
execute the network to simulate device behavior (Deshpande, sections 2-3 – teaches running the network model to model the behavior of the devices) and compute derivatives of the network with respect to selected inputs to minimize a cost function (Deshpande, section 4.3 – teaches minimizing a cost function).
While Deshpande teaches a network of sensor nodes and modeling the nodes based on an activation function, Deshpande does not explicitly teach an activation function/equation for each device/neuron.
Tulone teaches
embed one or more device-behavior equations for each device within the neuron as an activation function (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to modify Deshpande with the teachings of Tulone in order to improve efficiency in the field of heterogeneous sensor neural networks (Tulone, section 4.1 – “... For these reasons, it is more efficient for the sensor to compute m univariate AR models, although the multivariate model provides additional predictive power as it is able to capture correlations between measurements...”).
Regarding claim 50 (New), Deshpande in view of Tulone teaches all of the limitations of the system of claim 49 as noted above. Deshpande further teaches wherein the neurons comprise permanent inputs representing device properties (Deshpande, section 2 – teaches sensor attribute values) and temporary inputs representing state variables (Deshpande, section 4.1 – teaches edge values indicating packet probabilities [state values]) and cost function optimization selectively differentiates with respect to one or both input types (Deshpande, section 4.3 – teaches a cost function based on attributes and data transfers).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 49 above.
Regarding claim 51 (New), Deshpande in view of Tulone teaches all of the limitations of the system of claim 49 as noted above. Deshpande further teaches wherein each neuron corresponds to a physical device within a real-world system (Deshpande, section 2 – teaches physical sensors in a real-world system) and outputs of the neurons represent predicted physical behaviors of the devices (Deshpande, sections 3-4 – teaches outputting predicted sensor values to other sensors in the network).
It would have been obvious to one of ordinary skill in the art before the filing data of the claimed invention to combine the teachings of Deshpande and Tulone for the same reasons as disclosed in claim 49 above.
Regarding claim 52 (New), Deshpande in view of Tulone teaches all of the limitations of the system of claim 49 as noted above. Tulone further teaches wherein at least two neurons have activation functions representing unrelated physical processes (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to combine the teachings of Deshpande and Tulone in order to use attribute inputs to the equations to improve efficiency and capture data correlations (Tulone, section 4.1).
Regarding claim 53 (New), Deshpande teaches a non-transitory computer-readable medium storing instructions that, when executed by one or more processors (Deshpande, section 2 – teaches operating code [requiring memory] on a query processor), cause the processors to perform the method of
receiving, by a computing system (Deshpande, section 2 – teaches operating code [requiring memory] on a query processor), a linked network definition (Deshpande, section 2 – teaches modeling the behavior a sensor network with sensor nodes [neurons]) specifying:
a plurality of devices (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes),
for each device, one or more device-behavior equations that model physical behavior of the device (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes), and
creating, for each device a corresponding neuron in a neural network (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes);
embedding, within each neuron, the device-behavior equations associated with the corresponding device as an activation function of that neuron (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes);
connecting neurons according to relationships among the devices (Deshpande, sections 2-3 – teaches a network of sensors [neurons] to model the behavior of the sensors using an activation function; Deshpande, section 4 – teaches relationships between sensors nodes) so that:
inputs to each neuron correspond to outputs from nodes representing related devices (Deshpande, section 4.1 – teaches edges as weighted probabilities between nodes), and
outputs of each neuron correspond to transformed variables generated by evaluating its activation function (Deshpande, section 4.1 – teaches edges as weighted probabilities between nodes);
executing the neural network to compute values of variables by evaluating, within each neuron, the embedded device-behavior equations (Deshpande, sections 2-3 – teaches running the network model to model the behavior of the devices); and
computing a derivative of the neural network (Deshpande, sections 2-3 – teaches running the network model to model the behavior of the devices) to minimize a cost function that measures a difference between network output and real-world data (Deshpande, section 4.3 – teaches minimizing a cost function).
While Deshpande teaches a network of sensor nodes and modeling the nodes based on an activation function, Deshpande does not explicitly teach an activation function/equation for each device/neuron.
Tulone teaches
receiving, by a computing system, a linked network definition specifying:
a plurality of devices (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device),
for each device, one or more device-behavior equations that model physical behavior of the device (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device), and
embedding, within each neuron, the device-behavior equations associated with the corresponding device as an activation function of that neuron (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device);
connecting neurons according to relationships among the devices so that:
inputs to each neuron correspond to outputs from nodes representing related devices, and
outputs of each neuron correspond to transformed variables generated by evaluating its activation function (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device);
executing the neural network to compute values of variables by evaluating, within each neuron, the embedded device-behavior equations (Tulone, section 4.1 – teaches for each sensor device in the sensor network, creating either a multivariate model or multiple univariate models for the multiple attributes at each device).
It would have been obvious to one of ordinary skill in the art before the filing date of the claimed invention to modify Deshpande with the teachings of Tulone in order to improve efficiency in the field of heterogeneous sensor neural networks (Tulone, section 4.1 – “... For these reasons, it is more efficient for the sensor to compute m univariate AR models, although the multivariate model provides additional predictive power as it is able to capture correlations between measurements...”).
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 communication from the examiner should be directed to MARSHALL WERNER whose telephone number is (469) 295-9143. The examiner can normally be reached on Monday – Thursday 7:30 AM – 4:30 PM ET.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kamran Afshar, can be reached at (571) 272-7796. The fax 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.
/MARSHALL L WERNER/ Primary Examiner, Art Unit 2125
Prosecution Insights