DETAILED ACTION
This action is responsive to communications filed on December 14, 2023. This action is made Non-Final.
Claims 1-10 are pending in the case.
Claims 1, 9, and 10 are independent claims.
Claims 1-10 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 .
Priority
Receipt is acknowledged of papers submitted under 35 U.S.C. 119(a)-(d), which papers have been placed of record in the file.
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 10 is rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
Claim(s) 10:
Claim 10 recites a “computer-readable recording medium” for storing a program for executing a method. The recited “computer-readable recording medium” is not sufficiently limited to non-transitory media. Thus, the recited “computer-readable recording medium” is interpreted to include nonstatutory subject matter (e.g. signals, carrier waves, etc.).
Accordingly, Claim 10 fails to recite statutory subject matter under 35 U.S.C. 101.
Claims 1-4 and 6-10 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
With respect to claim 1:
2A Prong 1:
Claim 1 recites the following judicial exceptions:
normalizing the first physical data for the first fluid (mathematical concepts –as mathematical relationships, mathematical formulas or equations, or mathematical calculations (e.g. normalization of data is a mathematical concept including mathematical relationships, formulas or equations, and calculations).
based on the output data and the first label data corresponding to the first physical data for the first fluid; and acquiring a second label of second physical data for a second fluid ... to the second physical data based on acquiring the second physical data through the at least one sensor (mental process –can be performed in the human mind, or by a human using a pen and paper (e.g. a person may apply previous predictions or labels to new sensor data to determine new predictions or labels).
2A Prong 2: The additional elements recited in the claim do not integrate the judicial exception into a practical application.
Additional elements:
acquiring first physical data for a first fluid moving in a pipe through at least one sensor (mere instructions to apply the exception or implement the exception on a computer (e.g. a computer may collect or receive data from a sensor and perform preprocessing; see MPEP §2106.05(f).). The additional elements do not effectively integrate the abstract idea into a practical application. 2B: revisiting the additional elements, the additional elements do not amount to significantly more than the judicial exception – recited high level of generality and corresponds to storing and retrieving information in memory and performing calculations).
performing recurrent mapping on the normalized first physical data to generate at least one reservoir vector of an artificial neural network; updating a parameter between the at least one reservoir vector and data output through the at least one reservoir vector ... by applying the at least one reservoir vector (generally linking the use of a judicial exception to a particular technological environment or field of use (e.g. recurrent neural network may be used to process data, parameters may be update based on the processing, and predictions or labels may be generated; see MPEP §2106.05(h).). The additional elements do not effectively integrate the abstract idea into a practical application. 2B: revisiting the additional elements, the additional elements do not amount to significantly more than the judicial exception – recited high level of generality and corresponds to storing and retrieving information and performing calculations.).
With respect to claim 2:
2A Prong 2: The additional elements recited in the claim do not integrate the judicial exception into a practical application.
Additional elements:
wherein the at least one sensor includes a pressure gauge, flow meter, conductivity sensor, or impedance sensor, and the first physical data includes pressure, flow rate, conductivity, or impedance of the fluid (mere instructions to apply the exception or implement the exception on a computer (e.g. a computer may collect or receive data from a sensor and perform preprocessing; see MPEP §2106.05(f).). The additional elements do not effectively integrate the abstract idea into a practical application. 2B: revisiting the additional elements, the additional elements do not amount to significantly more than the judicial exception – recited high level of generality and corresponds to storing and retrieving information in memory and performing calculations).
With respect to claim 3:
2A Prong 1:
Claim 3 recites the following judicial exceptions:
wherein the normalizing comprises: reducing dimensionality of the first physical data by performing principal component analysis on the first physical data; acquiring a trend line by performing linear regression analysis on the first physical data of which the dimensionality has been reduced; removing a trend from the first physical data of which the dimensionality has been reduced based on the trend line and acquiring a standard deviation from the first physical data from which the trend has been removed; and normalizing the first physical data by dividing the first physical data by the standard deviation (mathematical concepts –as mathematical relationships, mathematical formulas or equations, or mathematical calculations (e.g. normalization of data is a mathematical concept including mathematical relationships, formulas or equations, and calculations).
With respect to claim 4:
2A Prong 1:
Claim 4 recites the following judicial exceptions:
wherein the first label data corresponding to the physical data is acquired by performing one-hot encoding on a label of the first physical data (mathematical concepts –as mathematical relationships, mathematical formulas or equations, or mathematical calculations (e.g. encoding of data is a mathematical concept including mathematical relationships, formulas or equations, and calculations).
With respect to claim 6:
2A Prong 1:
Claim 6 recites the following judicial exceptions:
acquiring second label data (mental process –can be performed in the human mind, or by a human using a pen and paper (e.g. a person may apply previous predictions or labels to new sensor data to determine new predictions or labels).
that is an average of the plurality of pieces of output data (mathematical concepts –as mathematical relationships, mathematical formulas or equations, or mathematical calculations (e.g. averaging of data is a mathematical concept including mathematical relationships, formulas or equations, and calculations).
2A Prong 2: The additional elements recited in the claim do not integrate the judicial exception into a practical application.
Additional elements:
acquiring the second physical data for the second fluid through the at least one sensor for a plurality of time units; (mere instructions to apply the exception or implement the exception on a computer (e.g. a computer may collect or receive data from a sensor and perform preprocessing; see MPEP §2106.05(f).). The additional elements do not effectively integrate the abstract idea into a practical application. 2B: revisiting the additional elements, the additional elements do not amount to significantly more than the judicial exception – recited high level of generality and corresponds to storing and retrieving information in memory and performing calculations).
acquiring a plurality of pieces of output data by applying the reservoir vector to each piece of the second physical data acquired for each of the plurality of time units (generally linking the use of a judicial exception to a particular technological environment or field of use (e.g. recurrent neural network may be used to process data, parameters may be update based on the processing, and predictions or labels may be generated; see MPEP §2106.05(h).). The additional elements do not effectively integrate the abstract idea into a practical application. 2B: revisiting the additional elements, the additional elements do not amount to significantly more than the judicial exception – recited high level of generality and corresponds to storing and retrieving information and performing calculations.).
With respect to claim 7:
2A Prong 2: The additional elements recited in the claim do not integrate the judicial exception into a practical application.
Additional elements:
wherein acquiring the second label comprising acquiring the second label by decoding the second label, and the second label contains information regarding physical properties o the second fluid (mere instructions to apply the exception or implement the exception on a computer (e.g. a computer may collect or receive data from a sensor and perform preprocessing and performing encoding/decoding; see MPEP §2106.05(f).). The additional elements do not effectively integrate the abstract idea into a practical application. 2B: revisiting the additional elements, the additional elements do not amount to significantly more than the judicial exception – recited high level of generality and corresponds to storing and retrieving information in memory and performing calculations).
With respect to claim 8:
2A Prong 2: The additional elements recited in the claim do not integrate the judicial exception into a practical application.
Additional elements:
wherein the updating comprises updating the parameter based on a Stochastic gradient descent (SGD) algorithm or an adaptive moment estimation (ADAM) algorithm, based on definition that a non-linear operation is performed on the at least one reservoir vector (generally linking the use of a judicial exception to a particular technological environment or field of use (e.g. recurrent neural network may be used to process data, parameters may be update based on the processing and using SGD, and predictions or labels may be generated; see MPEP §2106.05(h).). The additional elements do not effectively integrate the abstract idea into a practical application. 2B: revisiting the additional elements, the additional elements do not amount to significantly more than the judicial exception – recited high level of generality and corresponds to storing and retrieving information and performing calculations.).
With respect to claim 9:
Claim 9 further corresponds to claim 1 is rejection under the same rationale.
With respect to claim 10:
Claim 10 further corresponds to claim 1 is rejection under the same rationale.
2B continued: After considering all claim elements individually and as an ordered combination, it is determined that the claims do not include any additional elements that are sufficient to amount to significantly more than the judicial exception.
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.
Claim(s) 1, 5, 9, and 10 is/are rejected under 35 U.S.C. 103 as being unpatentable Peace et al., US Publication 2021/0372832 (“Peace”), in view of Uno, US Publication 2024/0386205 (“Uno), and further in view of Qingsong et al., US Publication 2021/0165770 (“Qingsong”).
Claim 1:
Peace teaches or suggests a method for processing and monitoring a flow signal in real time, performed by a device, the method comprising:
acquiring a first physical data for a first fluid moving in a pipe through at least one sensor (see Fig. 2, 3, and 7; para. 0010 -the neural network model may be pre-trained with data extracted from existing temperature sensors, together with actual flow data. The actual flow data may be obtained, for example, from traditional invasive flow sensors; para. 0060 - the neural network model is a model which has been pre-trained on reference data; para. 0063 - transfer of the raw environmental data to a cloud server, where it can be used improve trained neural network models; para. 0083 - multi-function sensor device that captures environmental data, and for example may be clipped onto the pipe 12 in which flow is to be measured; para. 0084 - Sensors 14 to 20 are all in contact with the pipe 12 through which flow is to be measure; para. 0108 - RNN may be pre-trained by providing samples of a multiplicity of sensor data to a neural network together with corresponding flow data. The flow data may be obtained using for example an impeller or ultrasonic based flow sensor and gathered from a large body of test sites; para. 0110 - pre-trained on data extracted from a large set of existing environmental sensors. The training process is performed in advance on a separate machine. In order to train the model, data is collected from environmental sensors together with actual flow data obtained from traditional invasive flow sensors.);
the first physical data for the first fluid and performing recurrent mapping on the ... first physical data to generate ... of an artificial neural network (see para. 0105 – Neural networks are capable of identifying non-linear patterns, such as temperature change to flow mapping; para. 0106 – applied to time-series data by use of a windowing mechanism and/or using a variation of the CNN known as a Recurrent Neural Network (RNN). The RNN is a neural network which has not only feedforward networks but also allows recurrent connections. In this way the network is able to refer to previous states and can therefore make use of previous inputs to model new outputs; para. 0108 - RNN may be pre-trained by providing samples of a multiplicity of sensor data to a neural network together with corresponding flow data. The flow data may be obtained using for example an impeller or ultrasonic based flow sensor and gathered from a large body of test sites; para. 0110 - pre-trained on data extracted from a large set of existing environmental sensors. The training process is performed in advance on a separate machine. In order to train the model, data is collected from environmental sensors together with actual flow data obtained from traditional invasive flow sensors. allows the RNN to learn how the environmental data relates to the actual flow data, and thus generate a model. The model is trained over time such that the outputs of the model match the actual flow data as closely as possible.);
updating a parameter ... based on the output data and first label data corresponding to the first physical data for the first fluid (see para. 0105 - multilayer feed-forward network, each layer of nodes receives inputs from the previous layers. The inputs to each node are combined using a weighted linear combination. The result is then modified by a nonlinear function before being output to the next layer. A backpropagation algorithm is used to calculate the weights in each layer; para. 0106 – applied to time-series data by use of a windowing mechanism and/or using a variation of the CNN known as a Recurrent Neural Network (RNN). The RNN is a neural network which has not only feedforward networks but also allows recurrent connections. In this way the network is able to refer to previous states and can therefore make use of previous inputs to model new outputs; para. 0108 - RNN may be pre-trained by providing samples of a multiplicity of sensor data to a neural network together with corresponding flow data. The flow data may be obtained using for example an impeller or ultrasonic based flow sensor and gathered from a large body of test sites; para. 0110 - pre-trained on data extracted from a large set of existing environmental sensors. The training process is performed in advance on a separate machine. In order to train the model, data is collected from environmental sensors together with actual flow data obtained from traditional invasive flow sensors. allows the RNN to learn how the environmental data relates to the actual flow data, and thus generate a model. The model is trained over time such that the outputs of the model match the actual flow data as closely as possible; para. 0119 - trained by providing samples of a multiplicity of sensor time-series data to the neural network together with corresponding flow time series data.); and
acquiring a second label of the second physical data for a second fluid by applying ... to the second physical data based on the acquiring the second physical data through the at least one sensor (see Fig. 2-5; para. 0009 – neural network model may allow the flow of fluid to be inferred from complex patterns of temperature change; para. 0014 - means for inferring the flow of fluid based on a change in temperature of the conduit over time, combined with one or more other environmental factors; para. 0083 - multi-function sensor device that captures environmental data, and for example may be clipped onto the pipe 12 in which flow is to be measured; para. 0084 - Sensors 14 to 20 are all in contact with the pipe 12 through which flow is to be measure; para. 0097 - able to use these parameters to infer water flow; para. 0104 - uses a trained deep recurrent neural network model in order to infer the flow of fluid in a pipe; para. 0107 - Recursive Neural Network (RNN), or more specifically a Long Short-Term Memory (LSTM) network, which retains a knowledge of previous states of the sensors, in order to infer a flow value from current sensor values. In the latter case, the algorithm is in the form of hidden nodes within a neural network; para. 0117 – retains a knowledge of previous states of the sensors in order to infer a flow value from current sensor values; para. 0127 - may use a neural network model such as that described above in order to infer the flow of fluid in the pipe.).
Peace does not appear to explicitly disclose normalizing ... normalized ... at least one reservoir vector; between the at least one reservoir vector and data output through the at least one reservoir vector.
Uno teaches or suggests at least one reservoir vector; between the at least one reservoir vector and data output through the at least one reservoir vector (see Fig. 4; para. 0061 - output sequence generation section uses, for example, an echo state network (ESN), which is one of recurrent neural network (RNN) models, to predict (generate) an output sequence; para. 0062 - enables input information to be linearly separable by mapping a low-dimensional vector of the input layer into a high-dimensional neuronal state space by non-linear transformation. ESN further reduces calculation cost in machine learning as compared with a common RNN; para. 0065 - transforms the input vector into a high dimensional feature vector by multiplying the input vector by a predetermined connection weight matrix. As described earlier, the high-dimensional feature vector has a higher dimension than a sum of a dimension of the token and a dimension of the context information vector; para. 0068 - output weight matrix multiplying section 83 multiplies a high-dimensional feature vector by an output weight matrix obtained by pre-training. the high dimensional feature vector is transformed into a lower dimensional output vector, and an output sequence is generated. component of the output weight matrix is found by pre-training and is stored, as the model parameter; para. 0069 - output sequence is generated by multiplying the high-dimensional feature vector by the output weight matrix obtained by pre-training; para. 0074 - reservoir layer 112, the input vector is multiplied by a connection weight matrix W and transformed into a high-dimensional feature vector; para. 0075 - high-dimensional feature vector is multiplied by an output weight matrix Wout so that an output vector is generated. A component of the matrix is set by machine learning which is carried out in advance. Machine learning of the output weight matrix is carried out with reference to, for example, training data including a plurality of sets.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include at least one reservoir vector; between the at least one reservoir vector and data output through the at least one reservoir vector for the purpose of efficiently processing and transforming an input using an echo state network, reducing machine learning calculation cost, as taught by Uno (0062 and 0069).
Qinsong further teaches or suggests normalizing ... normalized (see para. 0012 - further process the input time series data by performing normalization on the input time series data and/or removing outliers from the input time series data; para. 0037 - data processing (which may include, but is not limited to, data normalization, de-trending, outlier removal, etc.) on the input time series data to produce processed time series data. In implementations, the periodicity detection system 102 may apply a noise/outlier removal filter to remove noises and outliers from the input time series data.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include normalizing ... normalized for the purpose of efficiently preprocessing data to decrease variability, reducing further processing costs, as taught by Qingsong (0012 and 0037).
Claim(s) 9 and 10:
Claim(s) 9 and 10 correspond to Claim 1, and thus, Peace, Uno, and Qingsong teach or suggest the limitations of claim(s) 9 and 10 as well.
Claim 5:
Peace further teaches or suggests calculating an error between the data output ... and the first label data corresponding to the first physical data for the first fluid; and updating the parameter between ... the output data so that the calculated error is reduced (see para. 0105 – backpropagation algorithm is used to calculate the weights in each layer; LSTM network, is a recurrent neural network that is trained using Backpropagation.).
Uno further teaches or suggests through the at least one reservoir vector; the at least one reservoir vector (see Fig. 4; para. 0061 - output sequence generation section uses, for example, an echo state network (ESN), which is one of recurrent neural network (RNN) models, to predict (generate) an output sequence; para. 0062 - enables input information to be linearly separable by mapping a low-dimensional vector of the input layer into a high-dimensional neuronal state space by non-linear transformation. ESN further reduces calculation cost in machine learning as compared with a common RNN; para. 0065 - transforms the input vector into a high dimensional feature vector by multiplying the input vector by a predetermined connection weight matrix. As described earlier, the high-dimensional feature vector has a higher dimension than a sum of a dimension of the token and a dimension of the context information vector; para. 0068 - output weight matrix multiplying section 83 multiplies a high-dimensional feature vector by an output weight matrix obtained by pre-training. the high dimensional feature vector is transformed into a lower dimensional output vector, and an output sequence is generated. component of the output weight matrix is found by pre-training and is stored, as the model parameter; para. 0069 - output sequence is generated by multiplying the high-dimensional feature vector by the output weight matrix obtained by pre-training; para. 0074 - reservoir layer 112, the input vector is multiplied by a connection weight matrix W and transformed into a high-dimensional feature vector; para. 0075 - high-dimensional feature vector is multiplied by an output weight matrix Wout so that an output vector is generated. A component of the matrix is set by machine learning which is carried out in advance. Machine learning of the output weight matrix is carried out with reference to, for example, training data including a plurality of sets.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include through the at least one reservoir vector; the at least one reservoir vector for the purpose of efficiently processing and transforming an input using an echo state network, reducing machine learning calculation cost, as taught by Uno (0062 and 0069).
Claim(s) 2, 6, and 7 is/are rejected under 35 U.S.C. 103 as being unpatentable Peace, in view of Uno, in view of Qingsong, and further in view of Kusumura et al., US Publication 2018/0136076 (“Kusumura”).
Claim 2:
Kusumura further teaches or suggests wherein the at least one sensor includes a pressure gauge, flow meter, conductivity sensor, or impedance sensor, and the first physical data and the second physical data include pressure, flow rate, conductivity, or impedance of the fluid (see Fig. 1; para. 0033 - flow rate sensors including: a DMA flowmeter 14; and a flowmeter and a pressure regulating valve (PRV) 1. Measuring flow rates [liter/second] of water flowing into and flowing out from each of the monitor areas 12-1 to 12-7 through those flow rate sensors; para. 0039 - measured by the flow rate sensors 14 and 16; para. 0052 - learning unit 304 is configured to build, through learning, a prediction model (prediction equation of the water use amount) 207 for predicting normal values of the labeled data in response to the past environment condition data; para. 0069 – configured to receive the current flow rate.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include wherein the at least one sensor includes a pressure gauge, flow meter, conductivity sensor, or impedance sensor, and the first physical data and the second physical data include pressure, flow rate, conductivity, or impedance of the fluid for the purpose of efficiently collecting data for building prediction model related to flow rate, improving model generation, as taught by Kusumura (0033, 0052, and 0069).
Claim 6:
Peace further teaches or suggests acquiring the second physical data for the second fluid through the at least one sensor for a plurality of time units; acquiring a plurality of pieces of output data by applying ... to each piece of the second physical data acquired for each of the plurality of time units (see Fig. 2-5; para. 0009 – neural network model may allow the flow of fluid to be inferred from complex patterns of temperature change; para. 0014 - means for inferring the flow of fluid based on a change in temperature of the conduit over time, combined with one or more other environmental factors; para. 0083 - multi-function sensor device that captures environmental data, and for example may be clipped onto the pipe 12 in which flow is to be measured; para. 0084 - Sensors 14 to 20 are all in contact with the pipe 12 through which flow is to be measure; para. 0097 - able to use these parameters to infer water flow; para. 0104 - uses a trained deep recurrent neural network model in order to infer the flow of fluid in a pipe; para. 0107 - Recursive Neural Network (RNN), or more specifically a Long Short-Term Memory (LSTM) network, which retains a knowledge of previous states of the sensors, in order to infer a flow value from current sensor values. In the latter case, the algorithm is in the form of hidden nodes within a neural network; para. 0117 – retains a knowledge of previous states of the sensors in order to infer a flow value from current sensor values; para. 0127 - may use a neural network model such as that described above in order to infer the flow of fluid in the pipe.).
Uno further teaches or suggests the reservoir vector (see Fig. 4; para. 0061 - output sequence generation section uses, for example, an echo state network (ESN), which is one of recurrent neural network (RNN) models, to predict (generate) an output sequence; para. 0062 - enables input information to be linearly separable by mapping a low-dimensional vector of the input layer into a high-dimensional neuronal state space by non-linear transformation. ESN further reduces calculation cost in machine learning as compared with a common RNN; para. 0065 - transforms the input vector into a high dimensional feature vector by multiplying the input vector by a predetermined connection weight matrix. As described earlier, the high-dimensional feature vector has a higher dimension than a sum of a dimension of the token and a dimension of the context information vector; para. 0068 - output weight matrix multiplying section 83 multiplies a high-dimensional feature vector by an output weight matrix obtained by pre-training. the high dimensional feature vector is transformed into a lower dimensional output vector, and an output sequence is generated. component of the output weight matrix is found by pre-training and is stored, as the model parameter; para. 0069 - output sequence is generated by multiplying the high-dimensional feature vector by the output weight matrix obtained by pre-training; para. 0074 - reservoir layer 112, the input vector is multiplied by a connection weight matrix W and transformed into a high-dimensional feature vector; para. 0075 - high-dimensional feature vector is multiplied by an output weight matrix Wout so that an output vector is generated. A component of the matrix is set by machine learning which is carried out in advance. Machine learning of the output weight matrix is carried out with reference to, for example, training data including a plurality of sets.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include the reservoir vector for the purpose of efficiently processing and transforming an input using an echo state network, reducing machine learning calculation cost, as taught by Uno (0062 and 0069).
Kusumura further teaches or suggests acquiring second label data that is an average of the plurality of pieces of output data (see Fig. 1; para. 0013 - calculate an average value of the error values in the period of the window width defined by the first score parameter to estimate a water-leakage score representing a state of the water-leakage in the specific area; para. 0014 - calculating an average value of the error values in the period of the window width defined by the first score parameter to estimate a water-leakage score representing a state of the water-leakage in the specific area; para. 0033 - flow rate sensors including: a DMA flowmeter 14; and a flowmeter and a pressure regulating valve (PRV) 1. Measuring flow rates [liter/second] of water flowing into and flowing out from each of the monitor areas 12-1 to 12-7 through those flow rate sensors; para. 0039 - measured by the flow rate sensors 14 and 16; para. 0052 - learning unit 304 is configured to build, through learning, a prediction model (prediction equation of the water use amount) 207 for predicting normal values of the labeled data in response to the past environment condition data; para. 0069 – configured to receive the current flow rate; para. 0129 - calculate an average value of the error values in the period of the window width defined by the first score parameter to estimate a water-leakage score representing a state of the water-leakage in the specific area.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include acquiring second label data that is an average of the plurality of pieces of output data for the purpose of efficiently collecting data for building prediction model related to flow rate, improving model generation, as taught by Kusumura (0033, 0052, and 0069).
Claim 7:
Peace further teaches or suggests wherein the acquiring of the second label comprises acquiring the second label by decoding the second label data, and the second label data comprises information regarding physical properties of the second fluid (see Fig. 2-5; para. 0009 – neural network model may allow the flow of fluid to be inferred from complex patterns of temperature change; para. 0014 - means for inferring the flow of fluid based on a change in temperature of the conduit over time, combined with one or more other environmental factors; para. 0083 - multi-function sensor device that captures environmental data, and for example may be clipped onto the pipe 12 in which flow is to be measured; para. 0084 - Sensors 14 to 20 are all in contact with the pipe 12 through which flow is to be measure; para. 0097 - able to use these parameters to infer water flow; para. 0104 - uses a trained deep recurrent neural network model in order to infer the flow of fluid in a pipe; para. 0107 - Recursive Neural Network (RNN), or more specifically a Long Short-Term Memory (LSTM) network, which retains a knowledge of previous states of the sensors, in order to infer a flow value from current sensor values. In the latter case, the algorithm is in the form of hidden nodes within a neural network; para. 0117 – retains a knowledge of previous states of the sensors in order to infer a flow value from current sensor values; para. 0127 - may use a neural network model such as that described above in order to infer the flow of fluid in the pipe.).
Claim(s) 3 is/are rejected under 35 U.S.C. 103 as being unpatentable Peace, in view of Uno, in view of Qingsong, and further in view of Morris et al., US Publication 2021/0118547 (“Morris”).
Claim 3:
Qingsong further teaches or suggests reducing dimensionality of the first physical data by performing principal component analysis on the first physical data; acquiring a trend line by performing linear regression analysis on the first physical data of which the dimensionality has been reduced; removing a trend from the first physical data of which the dimensionality has been reduced based on the trend line (see para. 0012 - further process the input time series data by performing normalization on the input time series data and/or removing outliers from the input time series data; para. 0037 - data processing (which may include, but is not limited to, data normalization, de-trending, outlier removal, etc.) on the input time series data to produce processed time series data. In implementations, the periodicity detection system 102 may apply a noise/outlier removal filter to remove noises and outliers from the input time series data; para. 0040 - trend filter is said to be used as a de-trending filter and applied on the input time series data, in other instances, other de-trending filters may be used to estimate and remove the trend component from the input time series data. For example, the periodicity detection system 102 may apply a linear regression filter (or linear regression) on the input time series data to estimate the trending component, and remove the trending component from the input time series data;
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include reducing dimensionality of the first physical data by performing principal component analysis on the first physical data; acquiring a trend line by performing linear regression analysis on the first physical data of which the dimensionality has been reduced; removing a trend from the first physical data of which the dimensionality has been reduced based on the trend line and for the purpose of efficiently preprocessing data to decrease variability, reducing further processing costs, as taught by Qingsong (0012 and 0037).
Morris further teaches or suggests acquiring a standard deviation from the first physical data from which the trend has been removed; and normalizing the first physical data by dividing the first physical data by the standard deviation (see para. 0138 - which may be first normalized (subtract by mean and divided by standard deviation) and then concatenated into a high-dimensional regular time-series vector representation (X). Then, unsupervised learning is used to train an autoencoder (AE7) that can reconstruct the high-dimensional.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include acquiring a standard deviation from the first physical data from which the trend has been removed; and normalizing the first physical data by dividing the first physical data by the standard deviation for the purpose of efficiently preprocessing data to decrease variability, reducing further processing costs and improving training processes, as taught by Morris (0138).
Claim(s) 4 is/are rejected under 35 U.S.C. 103 as being unpatentable Peace, in view of Uno, in view of Qingsong, and further in view of Guzzonato et al., US Publication 2025/0069876 (“Guzzonato”).
Claim 4:
Guzzonato further teaches or suggests wherein the first label data corresponding to the first physical data is acquired by performing one-hot encoding on a label of the first physical data (see para. 0123 - measured intensities and sensor readbacks are
labelled by separating them into bins or classes; para. 0124 - data is prepared for training and validation by making two arrays, X and Y. X contains the features, and Y contains the labels/classes (assigned above). The labels are one-hot encoded. The data is split into training data, used to train a neural network, and testing data, used to test the neural network after training.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include wherein the first label data corresponding to the first physical data is acquired by performing one-hot encoding on a label of the first physical data for the purpose of efficiently preprocessing data to decrease variability, reducing further processing costs and improving training processes, as taught by Guzzonato (0124).
Claim(s) 8 is/are rejected under 35 U.S.C. 103 as being unpatentable Peace, in view of Uno, in view of Qingsong, and further in view of Yang, US Publication 2020/0255276 (“Yang”).
Claim 8:
Uno further teaches or suggests based on definition that a non-linear operation is performed on the at least one reservoir vector (see Fig. 4; para. 0061 - output sequence generation section uses, for example, an echo state network (ESN), which is one of recurrent neural network (RNN) models, to predict (generate) an output sequence; para. 0062 - enables input information to be linearly separable by mapping a low-dimensional vector of the input layer into a high-dimensional neuronal state space by non-linear transformation. ESN further reduces calculation cost in machine learning as compared with a common RNN; para. 0065 - transforms the input vector into a high dimensional feature vector by multiplying the input vector by a predetermined connection weight matrix. As described earlier, the high-dimensional feature vector has a higher dimension than a sum of a dimension of the token and a dimension of the context information vector; para. 0068 - output weight matrix multiplying section 83 multiplies a high-dimensional feature vector by an output weight matrix obtained by pre-training. the high dimensional feature vector is transformed into a lower dimensional output vector, and an output sequence is generated. component of the output weight matrix is found by pre-training and is stored, as the model parameter; para. 0069 - output sequence is generated by multiplying the high-dimensional feature vector by the output weight matrix obtained by pre-training; para. 0074 - reservoir layer 112, the input vector is multiplied by a connection weight matrix W and transformed into a high-dimensional feature vector; para. 0075 - high-dimensional feature vector is multiplied by an output weight matrix Wout so that an output vector is generated. A component of the matrix is set by machine learning which is carried out in advance. Machine learning of the output weight matrix is carried out with reference to, for example, training data including a plurality of sets.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include based on definition that a non-linear operation is performed on the at least one reservoir vector for the purpose of efficiently processing and transforming an input using an echo state network, reducing machine learning calculation cost, as taught by Uno (0062 and 0069).
Yang further teaches or suggests wherein the updating comprises updating the parameter based on a Stochastic gradient descent (SGD) algorithm or an adaptive moment estimation (ADAM) algorithm (see para. 0027 - Model parameters such as the convolution kernels and the weights of the fully connected layers are learned during training: typically a gradient based methods such as the stochastic gradient descent method or the mini-batch gradient descent method is used to gradually drive down a cross-entropy loss function, where parameter updates are propagated backwards with the back propagation algorithm.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Peace, to include wherein the updating comprises updating the parameter based on a Stochastic gradient descent (SGD) algorithm or an adaptive moment estimation (ADAM) algorithm for the purpose of efficiently reducing error using stochastic gradient descent and updating model parameters, improving model performance, as taught by Yang (0027).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Andrew T McIntosh whose telephone number is (571)270-7790. The examiner can normally be reached M-Th 8:00am-5:30pm.
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, Tamara Kyle can be reached at 571-272-4241. 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.
/ANDREW T MCINTOSH/Primary Examiner, Art Unit 2144
Prosecution Insights