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 .
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 07/14/2026 has been entered.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-3 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Cachapuz Santos Fontoura (hereinafter “CSF”) United States Patent Application Publication US 2024/0289839 in view of Lee United States Patent Application Publication US 2023/0214922.
Regarding claim 1, CSF discloses a computer-implemented method comprising:
generating an initial version of an analytics prediction machine learning model for predicting an analytics metric by learning initial parameters of the analytics prediction machine learning model utilizing model training data (CSF, para [0083], with regards to fig 5, element 506, training one or more machine learning models 115 on the one or more training datasets 220 via one or more LTR algorithms 226 based on one or more feature vectors (training feature vectors 304));
determining expected data channel contributions for the analytics metric according to prior observed data (CSF, para [0084], loss metric based on a comparison of the resultant ranking and known ranking and/or relevancy scores; CSF, para [0026], ranking based on input data provided by one or more data pipelines. Data pipeline interpreted as a channel)
…distributing or collecting data regarding digital content of a digital content campaign… (CSF, para [0020] One or more electronic advertisement rankings can be stored in one or more advertisement indexes accessible to an advertisement engine.)
generating, using a data channel contribution function and the initial parameters learned for the initial version of the analytics prediction machine learning model, predicted data channel contributions for the plurality of data channels (CSF, para [0084], the computer-implemented method 500 can include tuning (e.g., via the machine learning engine), by the system 100, one or more parameters of the one or more machine learning models 115 to minimize and/or reduce a loss metric, as defined by one or more loss functions, the tuning can be implemented across multiple iterations of the ranking operations); and
generating a modified analytics prediction machine learning model by iteratively updating the initial parameters until the updated parameters, as used in the data channel contribution function, produce updated predicted data channel contributions that are within a threshold similarity of the expected data channel contributions (CSF, para [0084], the computer-implemented method 500 can include tuning (e.g., via the machine learning engine), by the system 100, one or more parameters of the one or more machine learning models 115 to minimize and/or reduce a loss metric, as defined by one or more loss functions, the tuning can be implemented across multiple iterations of the ranking operations).
CSF does not disclose:
indicating, for a plurality of data channels comprising channels for distributing or collecting data …, what contributions to the analytics metric to expect from each data channel of the plurality of data channels;
Lee discloses:
indicating, for a plurality of data channels comprising channels for distributing or collecting data …, what contributions to the analytics metric to expect from each data channel of the plurality of data channels (Lee, para [0097], In the above description, the data in tensor form has been described as an example, which includes two channels corresponding to the 2D data for the ask price of a specific stock traded at a specific stock exchange and 2D data for the bid price of a specific stock traded at a specific stock exchange; Lee, para [0098], the data 800 in tensor form may further include one or more channels representing coordinate information. For example, the data 800 in tensor form may further include two channels corresponding to 2D data including values indicative of coordinates… Accordingly, the machine learning model that receives data in tensor form including one or more channels representing coordinate information may additionally use the information on the coordinate for the learning and inference, so as to output accurate prediction results based on accurate information on the position of each quantity data value.);
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified distributing or collecting data regarding digital content of a digital content campaign to include the steps of Lee. The motivation for doing so would have been to indicate what type of digital data would be received from each data channel (Lee, para [0098]).
Regarding claim 2, CSF discloses the computer-implemented method of claim 1. CSF additionally discloses wherein generating the initial version of the analytics prediction machine learning model comprises learning the initial parameters from the model training data that includes digital content campaign data indicating content distribution and resulting analytics metrics for one or more digital content campaigns (CSF, para [0027], advertisement analysis data including context information on the flight departure/destination locations, the aircraft, behavior data, etc.).
Regarding claim 3, CSF discloses the computer-implemented method of claim 1. CSF additionally discloses wherein determining the expected data channel contributions comprises accessing a database storing the prior observed data indicating respective contributions on impacting analytics metrics for a plurality of data channels (CSF, para [0026], ranking based on input data provided by one or more data pipelines; CSF, para [0027], rankings stored in index 104).
Regarding claim 15, CSF discloses a system comprising:
one or more memory devices comprising an analytics prediction machine learning model comprising parameters learned from an iterative training process that includes
generating an initial version of the analytics prediction machine learning model by learning initial parameters of the analytics prediction machine learning model (CSF, para [0083], with regards to fig 5, element 506, training one or more machine learning models 115 on the one or more training datasets 220 via one or more LTR algorithms 226 based on one or more feature vectors (training feature vectors 304));
generating, using a data channel contribution function and the initial parameters learned for the initial version of the analytics prediction machine learning model, predicted data channel contributions for a plurality of data channels (CSF, para [0084], the computer-implemented method 500 can include tuning (e.g., via the machine learning engine), by the system 100, one or more parameters of the one or more machine learning models 115 to minimize and/or reduce a loss metric, as defined by one or more loss functions, the tuning can be implemented across multiple iterations of the ranking operations); and
updating the initial parameters over multiple iterations until the parameters, as used in the data channel contribution function, produce updated predicted data channel contributions that are within a threshold similarity of expected data channel contributions (CSF, para [0084], to minimize and/or reduce a loss metric, as defined by one or more loss functions, the tuning can be implemented across multiple iterations of the ranking operations)
one or more processors configured to cause the system to:
access content distribution data for a digital content campaign (CSF, para [0027], accesses data stored in index 104);
determine a target analytics metric for the digital content campaign (CSF, para [0084], loss metric based on a comparison of the resultant ranking and known ranking and/or relevancy scores; CSF, para [0026], ranking based on input data provided by one or more data pipelines. Data pipeline interpreted as a channel); and
generate an analytics prediction for the target analytics metric utilizing the analytics prediction machine learning model to process the content distribution data according to the parameters learned from the iterative training process (CSF, para [0027], As shown in FIG. 1, the one or more data pipelines 112 can be operably coupled to the one or more advertisement analysis devices 102, which can apply one or more machine learning engines 114 to rank available advertisements… advertisement analysis data including context information on the flight departure/destination locations, the aircraft, behavior data, etc. … the data supplied by the one or more data pipelines 112 can be used as input data for the trained machine learning model 115 to implement one or more advertisement ranking operations… the one or more advertisement analysis devices 102 can generate one or more electronic advertisement rankings based on the advertisements' relevance to one or more feature vectors extracted from the input data).
CSF does not disclose:
for an analytics metric according to prior observed data indicating what contributions to the analytics metric to expect from each data channel of the plurality of data channels
Lee discloses:
for an analytics metric according to prior observed data indicating what contributions to the analytics metric to expect from each data channel of the plurality of data channels (Lee, para [0097], In the above description, the data in tensor form has been described as an example, which includes two channels corresponding to the 2D data for the ask price of a specific stock traded at a specific stock exchange and 2D data for the bid price of a specific stock traded at a specific stock exchange; Lee, para [0098], the data 800 in tensor form may further include one or more channels representing coordinate information. For example, the data 800 in tensor form may further include two channels corresponding to 2D data including values indicative of coordinates… Accordingly, the machine learning model that receives data in tensor form including one or more channels representing coordinate information may additionally use the information on the coordinate for the learning and inference, so as to output accurate prediction results based on accurate information on the position of each quantity data value.)
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified distributing or collecting prior observed data to include the steps of Lee. The motivation for doing so would have been to indicate what type of data would be received from each data channel (Lee, para [0098]).
Claim Rejections - 35 USC § 103
Claims 4-5, 8-12, 17-19 are rejected under 35 U.S.C. 103 as being unpatentable over Cachapuz Santos Fontoura (hereinafter “CSF”) United States Patent Application Publication US 2024/0289839 in view of Lee United States Patent Application Publication US 2023/0214922, and further in view of Zagayevskiy United States Patent Application Publication US 2021/0133375.
Regarding claim 4, CSF discloses the computer-implemented method of claim 1. CSF discloses wherein generating the modified analytics prediction machine learning model by iteratively updating the initial parameters comprises, for a number of iterations:
comparing a point with an additional point representing the expected data channel contributions (CSF, para [0084], to minimize and/or reduce a loss metric, as defined by one or more loss functions, the tuning can be implemented across multiple iterations of the ranking operations).
CSF does not disclose:
generating the updated parameters from the initial parameters and the expected data channel contributions;
generating a point in parameter space representing the updated parameters utilizing the data channel contribution function; and
comparing the point in the parameter space with an additional point in the parameter space.
Zagayevskiy discloses:
generating updated parameters from the expected data channel contributions;
generating a point in parameter space representing the updated parameters utilizing the data channel contribution function; and
comparing the point in the parameter space with an additional point in the parameter space (Zagayevskiy, para [0030, 32], with regards to fig 3, illustrated is a diagram of an algorithm 160 for the flow simulator module 48 of the analytics platform 12 for generating reservoir management workflows and forecasts from a high-dimensional parameter data space; fig. 3, element 170, updates parameters with output variables and algorithmic models until the criteria is satisfied).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the updating of parameters to include the steps of Zagayevskiy. The motivation for doing so would have been to provide less processing with an acceptable degree of accuracy (Zagayevskiy, para [0002]).
Regarding claim 5, CSF discloses the computer-implemented method of claim 1. CSF does not disclose the additional limitations of the present claim.
Zagayevskiy discloses wherein generating the modified analytics prediction machine learning model comprises iteratively updating the initial parameters according to an objective function that incorporates the expected data channel contributions, the predicted data channel contributions, an observed analytics metric, and a predicted analytics metric generated according to the initial parameters (Zagayevskiy, para [0015, 17, 25], The machine learning and training and validation module 74 can iteratively generate reduced parameter space algorithmic models from the computationally complex algorithmic model using at least one training dataset and at least one validation dataset, i.e. reservoir models, and at least one selected from a group of: input variables 56, output variables 58, updated reservoir model 76, history matching input variables 78, the optimized reservoir model 84, and optimization input variables 86).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the updating of parameters to include the steps of Zagayevskiy. The motivation for doing so would have been to provide less processing with an acceptable degree of accuracy (Zagayevskiy, para [0002]).
Regarding claim 8, CSF discloses a non-transitory computer readable medium storing executable instructions which, when executed by a processing device, cause the processing device to perform operations comprising:
generating an initial version of an analytics prediction machine learning model for predicting an analytics metric by learning initial parameters of the analytics prediction machine learning model utilizing model training data (CSF, para [0083], with regards to fig 5, element 506, training one or more machine learning models 115 on the one or more training datasets 220 via one or more LTR algorithms 226 based on one or more feature vectors (training feature vectors 304)); and
determining expected data channel contributions for the analytics metric according to prior observed data (CSF, para [0084], loss metric based on a comparison of the resultant ranking and known ranking and/or relevancy scores; CSF, para [0026], ranking based on input data provided by one or more data pipelines. Data pipeline interpreted as a channel)
…distributing or collecting data regarding digital content of a digital content campaign… (CSF, para [0020] One or more electronic advertisement rankings can be stored in one or more advertisement indexes accessible to an advertisement engine.)
generating, using a data channel contribution function and the initial parameters learned for the initial version of the analytics prediction machine learning model, predicted data channel contributions for the plurality of data channels (CSF, para [0084], the computer-implemented method 500 can include tuning (e.g., via the machine learning engine), by the system 100, one or more parameters of the one or more machine learning models 115 to minimize and/or reduce a loss metric, as defined by one or more loss functions, the tuning can be implemented across multiple iterations of the ranking operations); and
generating a modified analytics prediction machine learning model by iteratively updating the initial parameters (CSF, para [0084], to minimize and/or reduce a loss metric, as defined by one or more loss functions, the tuning can be implemented across multiple iterations of the ranking operations).
CSF does not disclose:
indicating, for a plurality of data channels comprising channels for distributing or collecting data …, what contributions to the analytics metric to expect from each data channel of the plurality of data channels;
generating a modified analytics prediction machine learning model by iteratively: generating updated parameters from the initial parameters and expected data channel contributions; generating a point in parameter space representing the updated parameters utilizing the data channel contribution function; and comparing the point in parameter space with an additional point in the parameter space representing the expected data channel contributions.
Lee discloses:
indicating, for a plurality of data channels comprising channels for distributing or collecting data …, what contributions to the analytics metric to expect from each data channel of the plurality of data channels (Lee, para [0097], In the above description, the data in tensor form has been described as an example, which includes two channels corresponding to the 2D data for the ask price of a specific stock traded at a specific stock exchange and 2D data for the bid price of a specific stock traded at a specific stock exchange; Lee, para [0098], the data 800 in tensor form may further include one or more channels representing coordinate information. For example, the data 800 in tensor form may further include two channels corresponding to 2D data including values indicative of coordinates… Accordingly, the machine learning model that receives data in tensor form including one or more channels representing coordinate information may additionally use the information on the coordinate for the learning and inference, so as to output accurate prediction results based on accurate information on the position of each quantity data value.);
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified distributing or collecting data regarding digital content of a digital content campaign to include the steps of Lee. The motivation for doing so would have been to indicate what type of digital data would be received from each data channel (Lee, para [0098]).
Zagayevskiy discloses:
generating a modified analytics prediction machine learning model by iteratively: generating updated parameters from the expected data channel contributions; generating a point in parameter space representing the updated parameters utilizing a data channel contribution function; and comparing the point in parameter space with an additional point in the parameter space representing the expected data channel contributions (Zagayevskiy, para [0030, 32], with regards to fig 3, illustrated is a diagram of an algorithm 160 for the flow simulator module 48 of the analytics platform 12 for generating reservoir management workflows and forecasts from a high-dimensional parameter data space; fig. 3, element 170, updates parameters with output variables and algorithmic models until the criteria is satisfied).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the updating of parameters to include the steps of Zagayevskiy. The motivation for doing so would have been to provide less processing with an acceptable degree of accuracy (Zagayevskiy, para [0002]).
Regarding claim 9, CSF in view of Zagayevskiy discloses the non-transitory computer readable medium of claim 8. CSF additionally discloses wherein generating the initial version of the analytics prediction machine learning model comprises learning the initial parameters from the model training data that includes digital content campaign data indicating content distribution and corresponding analytics metrics for one or more digital content campaigns (CSF, para [0027], advertisement analysis data including context information on the flight departure/destination locations, the aircraft, behavior data, etc.).
Regarding claim 10, CSF in view of Zagayevskiy discloses the non-transitory computer readable medium of claim 8. CSF additionally discloses wherein determining the expected data channel contributions comprises determining the expected data channel contributions from the prior observed data indicating, for a plurality of data channels, respective contributions on impacting the analytics metric (CSF, para [0026], ranking based on input data provided by one or more data pipelines; CSF, para [0027], rankings stored in index 104).
Regarding claim 11, CSF in view of Zagayevskiy discloses the non-transitory computer readable medium of claim 8. CSF additionally discloses wherein generating the modified analytics prediction machine learning model comprises iteratively updating the initial parameters, generating the point in the parameter space, and comparing the point in the parameter space with the additional point until the point and the additional point are within a threshold distance of each other in the parameter space (CSF, para [0084], to minimize and/or reduce a loss metric, as defined by one or more loss functions, the tuning can be implemented across multiple iterations of the ranking operations).
Regarding claim 12, CSF in view of Zagayevskiy discloses the non-transitory computer readable medium of claim 8. Zagayevskiy additionally discloses wherein generating the modified analytics prediction machine learning model comprises iteratively updating the initial parameters according to an objective function that incorporates the expected data channel contributions, the predicted data channel contributions, an observed analytics metric, and a predicted analytics metric generated by a previous version of the analytics prediction machine learning model according to a previous version of parameters (Zagayevskiy, para [0015, 17, 25], The machine learning and training and validation module 74 can iteratively generate reduced parameter space algorithmic models from the computationally complex algorithmic model using at least one training dataset and at least one validation dataset, i.e. reservoir models, and at least one selected from a group of: input variables 56, output variables 58, updated reservoir model 76, history matching input variables 78, the optimized reservoir model 84, and optimization input variables 86).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the updating of parameters to include the steps of Zagayevskiy. The motivation for doing so would have been to provide less processing with an acceptable degree of accuracy (Zagayevskiy, para [0002]).
Regarding claim 17, CSF discloses the system of claim 15. CSF does not disclose the additional limitations of the present claim.
Zagayevskiy discloses wherein the analytics prediction machine learning model comprises parameters learned by iteratively: generating the updated parameters from the expected data channel contributions; generating a point in parameter space representing the updated parameters utilizing the data channel contribution function; and comparing the point in the parameter space with an additional point in the parameter space representing the expected data channel contributions (Zagayevskiy, para [0030, 32], with regards to fig 3, illustrated is a diagram of an algorithm 160 for the flow simulator module 48 of the analytics platform 12 for generating reservoir management workflows and forecasts from a high-dimensional parameter data space; fig. 3, element 170, updates parameters with output variables and algorithmic models until the criteria is satisfied).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the updating of parameters to include the steps of Zagayevskiy. The motivation for doing so would have been to provide less processing with an acceptable degree of accuracy (Zagayevskiy, para [0002]).
Regarding claim 18, CSF discloses the system of claim 15. CSF does not disclose the additional limitations of the present claim.
Zagayevskiy discloses wherein the analytics prediction machine learning model comprises parameters learned according to an objective function that incorporates the expected data channel contributions, predicted data channel contributions, an observed analytics metric, and a predicted analytics metric (Zagayevskiy, para [0015, 17, 25], The machine learning and training and validation module 74 can iteratively generate reduced parameter space algorithmic models from the computationally complex algorithmic model using at least one training dataset and at least one validation dataset, i.e. reservoir models, and at least one selected from a group of: input variables 56, output variables 58, updated reservoir model 76, history matching input variables 78, the optimized reservoir model 84, and optimization input variables 86).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the updating of parameters to include the steps of Zagayevskiy. The motivation for doing so would have been to provide less processing with an acceptable degree of accuracy (Zagayevskiy, para [0002]).
Regarding claim 19, CSF in view of Zagayevskiy discloses the system of claim 18. CSF does not disclose the additional limitations of the present claim.
Zagayevskiy discloses wherein the objective function produces parameters for the analytics prediction machine learning model that reduce a difference between predicted data channel contributions and the expected data channel contributions (Zagayevskiy, para [0030, 32], with regards to fig 3, illustrated is a diagram of an algorithm 160 for the flow simulator module 48 of the analytics platform 12 for generating reservoir management workflows and forecasts from a high-dimensional parameter data space; fig. 3, element 170, updates parameters with output variables and algorithmic models until the criteria is satisfied).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the updating of parameters to include the steps of Zagayevskiy. The motivation for doing so would have been to provide less processing with an acceptable degree of accuracy (Zagayevskiy, para [0002]).
Claim Rejections - 35 USC § 103
Claims 6-7 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Cachapuz Santos Fontoura (hereinafter “CSF”) United States Patent Application Publication US 2024/0289839 in view of Lee United States Patent Application Publication US 2023/0214922, and further in view of Pyzer-Knapp United States Patent Application Publication US 2022/0128972.
Regarding claim 6, CSF discloses the computer-implemented method of claim 1. CSF does not disclose the additional limitations of the present claim.
Pyzer-Knapp discloses wherein generating the modified analytics prediction machine learning model comprises utilizing a surrogate function to modify terms of the data channel contribution function for iteratively updating the initial parameters (Pyzer-Knapp, para [0017, 24, 29], iteratively updates surrogate model).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the steps to include using a surrogate function based on the teachings of Pyzer-Knapp. The motivation for doing so would have been to implement a Bayesian optimization because of its strong performance of optimization (Pyzer-Knapp, para [0002]).
Regarding claim 7, CSF in view of Pyzer-Knapp discloses the computer-implemented method of claim 6. Pyzer-Knapp additionally discloses wherein utilizing the surrogate function as part of updating the initial parameters of the analytics prediction machine learning model comprises replacing predicted analytics metrics of the data channel contribution function with observed analytics metrics (Pyzer-Knapp, para [0017, 33], new sample and the corresponding output (observation) is then used to update the surrogate model).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the steps to include using a surrogate function based on the teachings of Pyzer-Knapp. The motivation for doing so would have been to implement a Bayesian optimization because of its strong performance of optimization (Pyzer-Knapp, para [0002]).
Regarding claim 20, CSF discloses the system of claim 18. CSF does not disclose the additional limitations of the present claim.
Pyzer-Knapp discloses wherein the objective function comprises a surrogate function that substitutes observed analytics metrics for predicted analytics metrics as a component of the objective function (Pyzer-Knapp, para [0017, 24, 29], iteratively updates surrogate model).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the steps to include using a surrogate function based on the teachings of Pyzer-Knapp. The motivation for doing so would have been to implement a Bayesian optimization because of its strong performance of optimization (Pyzer-Knapp, para [0002]).
Claim Rejections - 35 USC § 103
Claim(s) 13-14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Cachapuz Santos Fontoura (hereinafter “CSF”) United States Patent Application Publication US 2024/0289839 in view of Lee United States Patent Application Publication US 2023/0214922, further in view of Zagayevskiy United States Patent Application Publication US 2021/0133375, and further in view of Pyzer-Knapp United States Patent Application Publication US 2022/0128972.
Regarding claim 13, CSF in view of Zagayevskiy discloses the non-transitory computer readable medium of claim 8. CSF in view of Zagayevskiy does not disclose the additional limitations of the present claim.
Pyzer-Knapp discloses wherein generating the modified analytics prediction machine learning model comprises utilizing a surrogate function to iteratively update the initial parameters (Pyzer-Knapp, para [0017, 24, 29], iteratively updates surrogate model).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the steps to include using a surrogate function based on the teachings of Pyzer-Knapp. The motivation for doing so would have been to implement a Bayesian optimization because of its strong performance of optimization (Pyzer-Knapp, para [0002]).
Regarding claim 14, CSF in view of Zagayevskiy in further view of Pyzer-Knapp discloses tThe non-transitory computer readable medium of claim 13. Pyzer-Knapp additionally discloses wherein utilizing the surrogate function as part of updating the initial parameters of the analytics prediction machine learning model comprises: determining a modified data channel contribution function by replacing predicted analytics metrics within the data channel contribution function with observed analytics metrics; and generating the surrogate function to substitute for an objective function designed for updating the initial parameters of the analytics prediction machine learning model by utilizing the modified data channel contribution function (Pyzer-Knapp, para [0017, 33], new sample and the corresponding output (observation) is then used to update the surrogate model).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the steps to include using a surrogate function based on the teachings of Pyzer-Knapp. The motivation for doing so would have been to implement a Bayesian optimization because of its strong performance of optimization (Pyzer-Knapp, para [0002]).
Claim Rejections - 35 USC § 103
Claim(s) 16 is rejected under 35 U.S.C. 103 as being unpatentable over Cachapuz Santos Fontoura (hereinafter “CSF”) United States Patent Application Publication US 2024/0289839 in view of Lee United States Patent Application Publication US 2023/0214922, and further in view of Chang United States Patent US 10,346,870.
Regarding claim 16, CSF discloses the system of claim 15. CSF does not disclose the additional limitations of the present claim.
Chang discloses wherein generating the analytics prediction for the target analytics metric comprises utilizing the analytics prediction machine learning model to generate a predicted conversion rate for the digital content campaign from the content distribution data (Chang, col 3, rows 47-61, in addition text corresponding to fig 3 element 307, fig 7A element 706, fig 7B element 714-715, probability that a consumer will accept an offer for the promotion program represents a conversion rate. Generates prediction based on collected performance data for a time).
Before the time of the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have modified the system to include the promotional analytics of Chang. The motivation for doing so would have been to have a higher likelihood of acceptance for the consumer (Chang, col 1, rows 18-26).
Response to Arguments
For claims 1 and 15, Lee teaches indicating for a plurality of data channels comprising channels for distributing or collecting data, what contributions to the analytics metric to expect from each data channel according to paragraphs 0097 and 0098, where tensors with data channels are provided to a machine learning model, and the data channels are indicated to contain 2D coordinate data of stock prices, which are the analytics metric contributions. CSF teaches distributing or collecting data regarding digital content of a digital content campaign in paragraph 0020. The teachings of CSF and Lee could be combined to indicate what contributions to the analytics metric to expect from each data channel containing digital content.
For claim 1 and 15, CSF teaches generating predicted data channel contributions for the plurality of data channels using a data channel contribution function and initial parameters learned for the initial version of the analytics prediction machine learning model in paragraph 0084. The predicted data channel contributions are the loss metrics that are being minimized and/or reduced. The data channel contribution function are the one or more loss functions 308. The initial parameters are the one or more parameters that are going to be tuned by the system 100.
For claim 1, CSF also teaches that a modified analytics prediction machine learning model is generated by tuning one or more parameters in paragraph 0084. The tuned model represents the modified portion of the machine learning model while the one or more parameters that are tuned represent the iteratively updated parameters. The loss functions are the data channel contribution functions that produce predicted contributions, and the minimization or reduction of a loss function’s loss metric is what produces predicted contributions that are within a threshold similarity of expected data channel contributions.
For claim 15, CSF teaches utilizing the analytics prediction ML model in paragraph 0027. The advertisement analysis devices, which can apply one or more machine learning engines, analyze the analytics metrics by ranking them. The data supplied by the pipelines, which are extracted for their feature vectors, are the parameters learned from the training process to assist with ranking.
For claim 15, CSF teaches generating an initial version of the analytics prediction machine learning model by learning initial parameters of the analytics prediction machine learning model in paragraph 0083. This is similar to the first generating step of claim 1, where the one or more machine learning models 115 is trained on the training datasets 220 using LTR algorithms 226 based on one or more training feature vectors 304. The analytics prediction machine learning model are the one or more machine learning models 115, and the training feature vectors 304 are the initial parameters of the analytics prediction machine learning model.
For claim 8, Lee teaches indicating for a plurality of data channels comprising channels for distributing or collecting data, what contributions to the analytics metric to expect from each data channel according to paragraphs 0097 and 0098, where tensors with data channels are provided to a machine learning model, and the data channels are indicated to contain 2D coordinate data of stock prices, which are the analytics metric contributions. CSF teaches distributing or collecting data regarding digital content of a digital content campaign in paragraph 0020. The teachings of CSF and Lee could be combined to indicate what contributions to the analytics metric to expect from each data channel containing digital content.
Applicant's arguments regarding the 35 USC 103 rejection are moot in view of the new grounds of rejection necessitated by applicant's amendments.
With respect to the arguments for the 35 U.S.C. 103 rejections from pg. 13-16 that CSF and Zagayevskiy do not teach the newly amended limitations because they fail to teach the currently amended independent claims 1, 8, and 15, the examiner disagrees. As explained in the rejection above, CSF teaches all of the limitations in the currently amended independent claims 1, 8, and 15. The arguments regarding Zagayevskiy are not relevant as Zagayevskiy is not being used for the newly added limitations.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DAVID KIM whose telephone number is (571)272-4331. The examiner can normally be reached 7:30 AM - 4:30 PM.
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, Matthew Ell can be reached at (571) 270-3264. 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.
/D.K./Examiner, Art Unit 2141
/MATTHEW ELL/Supervisory Patent Examiner, Art Unit 2141
Prosecution Insights