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 .
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding Claim 1,
Claim 1 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 1 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
map an input context of an ad opportunity to an output label vector having a plurality of probabilities, each of which predicts a likelihood of a reaction of a corresponding one of the plurality of audiences to the input context of the ad opportunity
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) and/or mathematical concepts (mathematical relationships, mathematical formulas or equations, mathematical calculations) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
map an input context of an ad opportunity to an output label vector having a plurality of probabilities, each of which predicts a likelihood of a reaction of a corresponding one of the plurality of audiences to the input context of the ad opportunity (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional elements of “at least one processor, a memory, and a communication platform for predictive targeting”, “initializing a joint predictive model with initial model parameters” and “the learned joint predictive model is to be used”, as drafted, amount to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Moreover, the claim recites additional element(s) that amount to insignificant extra-solution activities, which do not integrate a judicial exception into a practical application. For example, the additional element of “obtaining training data based on cookieless online traffic, wherein the training data comprises pairs of data, each of the pairs including an ad opportunity context corresponding to an ad served to a corresponding one of a plurality of audiences and a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served, of a corresponding one of the plurality of audiences in the ad opportunity context” amounts to mere data gathering, which is an insignificant extra-solution activity that does not integrate a judicial exception into a practical application. See MPEP 2106.05(g). Further, the additional element of “machine learning, based on the training data, model parameters of the joint predictive model based on the initial model parameters by minimizing a loss in an iterative process” recites an insignificant extra-solution activity which does not integrate a judicial exception into a practical application. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Moreover, the additional element of “obtaining training data based on cookieless online traffic, wherein the training data comprises pairs of data, each of the pairs including an ad opportunity context corresponding to an ad served to a corresponding one of a plurality of audiences and a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served, of a corresponding one of the plurality of audiences in the ad opportunity context” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See MPEP 2106.05(d)(II) “The courts have recognized the following computer functions as well‐understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity… i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Intellectual Ventures v. Symantec, 838 F.3d 1307, 1321; 120 USPQ2d 1353, 1362 (Fed. Cir. 2016)”. Further, the additional element of “machine learning, based on the training data, model parameters of the joint predictive model based on the initial model parameters by minimizing a loss in an iterative process” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See Abraham et al. (US 20230368915 A1 (Filed 2021)) Specification [0070] “Conventional machine learning model training system 100 can train the machine learning model 110 to minimize the (cumulative) loss function 130 by performing multiple iterations of conventional machine learning model training techniques on training data items”. Therefore, the claim is not patent eligible.
Regarding Claim 2,
Claim 2 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 2 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The following limitations:
the reaction includes one of a conversion and a lack of conversion
a label in the label vector indicates whether the reaction is a conversion or a lack of conversion
as drafted, under the broadest reasonable interpretation, cover mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
the reaction includes one of a conversion and a lack of conversion (corresponds to evaluation and judgment);
a label in the label vector indicates whether the reaction is a conversion or a lack of conversion (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: See corresponding analysis of claim 1.
Step 2B Analysis: See corresponding analysis of claim 1.
Regarding Claim 3,
Claim 3 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 3 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
an ad opportunity context is characterized based on a plurality of contextual features
each contextual feature is encoded via a first representation of a first dimension
each first representation is characterized by a feature vector of a second dimension, wherein the first dimension is larger than the second dimension
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
an ad opportunity context is characterized based on a plurality of contextual features (corresponds to evaluation and judgment);
each contextual feature is encoded via a first representation of a first dimension (corresponds to evaluation and judgment);
each first representation is characterized by a feature vector of a second dimension, wherein the first dimension is larger than the second dimension (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: See corresponding analysis of claim 1.
Step 2B Analysis: See corresponding analysis of claim 1.
Regarding Claim 4,
Claim 4 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 4 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
predict an output label vector
representing each contextual feature of an ad opportunity context via the first representation weighted by a first coefficient
[representing] each first representation via the feature vector
incorporating interactions between feature vectors of different contextual features of each ad opportunity context weighted by a second set of coefficients
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
predict an output label vector (corresponds to evaluation and judgment);
representing each contextual feature of an ad opportunity context via the first representation weighted by a first coefficient (corresponds to evaluation and judgment);
[representing] each first representation via the feature vector (corresponds to evaluation and judgment);
incorporating interactions between feature vectors of different contextual features of each ad opportunity context weighted by a second set of coefficients (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “the joint predictive model is constructed”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Therefore, the claim is not patent eligible.
Regarding Claim 5,
Claim 5 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 5 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: See corresponding analysis of claim 4.
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “initializing values of feature vectors related to the plurality of contextual features”, “initializing values of the first coefficients used to weigh the corresponding plurality of contextual features”, and “initializing values of the second set of coefficients used to weigh the interactions”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Therefore, the claim is not patent eligible.
Regarding Claim 6,
Claim 6 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 6 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
predicting an output label vector with a plurality of probabilities based on current model parameters of the joint predictive model, wherein each probability of the output label vector indicates a likelihood of a reaction from a corresponding one of the plurality of audiences
computing a loss based on the predicted output label vector and the label vector from the pair of data from the training data
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) and/or mathematical concepts (mathematical relationships, mathematical formulas or equations, mathematical calculations) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
predicting an output label vector with a plurality of probabilities based on current model parameters of the joint predictive model, wherein each probability of the output label vector indicates a likelihood of a reaction from a corresponding one of the plurality of audiences (corresponds to evaluation and judgment);
computing a loss based on the predicted output label vector and the label vector from the pair of data from the training data (corresponds to mathematical calculations).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “the steps of obtaining, predicting, computing, and adjusting are repeated in the iterative process until the loss satisfies a pre-determined criterion to generate the joint predictive model with converged model parameters”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Further, the additional element of “adjusting the current model parameters of the joint predictive model by minimizing the loss” recites an insignificant extra-solution activity which does not integrate a judicial exception into a practical application. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Further, the additional element of “adjusting the current model parameters of the joint predictive model by minimizing the loss” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See Abraham et al. (US 20230368915 A1 (Filed 2021)) Specification [0070] “Conventional machine learning model training system 100 can train the machine learning model 110 to minimize the (cumulative) loss function 130 by performing multiple iterations of conventional machine learning model training techniques on training data items”. Therefore, the claim is not patent eligible.
Regarding Claim 7,
Claim 7 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 7 is directed to a method, which is directed to a process, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
creating first representations for corresponding contextual features of the input context of the ad opportunity
generating an output label vector with respect to the plurality of audiences, based on the joint predictive model with converged model parameters, to predict probabilities of reactions of the respective plurality of audiences to the ad opportunity context
select one or more of the plurality of audiences based on the predicted probabilities in the output label vector
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
creating first representations for corresponding contextual features of the input context of the ad opportunity (corresponds to evaluation and judgment);
generating an output label vector with respect to the plurality of audiences, based on the joint predictive model with converged model parameters, to predict probabilities of reactions of the respective plurality of audiences to the ad opportunity context (corresponds to evaluation and judgment);
select one or more of the plurality of audiences based on the predicted probabilities in the output label vector (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “to enable the DSP to”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Moreover, the claim recites additional element(s) that amount to insignificant extra-solution activities, which do not integrate a judicial exception into a practical application. For example, the additional element of “receiving, from a demand side platform (DSP), an input context of an ad opportunity” and “transmitting the output label vector to the DSP” amounts to mere data gathering, which is an insignificant extra-solution activity that does not integrate a judicial exception into a practical application. See MPEP 2106.05(g). Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Moreover, the additional element of “obtaining training data comprising pairs of data, each of the pairs includes an ad opportunity context corresponding to an ad served to a plurality of audiences and a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served, of a corresponding one of the plurality of audiences in the ad opportunity context” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See MPEP 2106.05(d)(II) “The courts have recognized the following computer functions as well‐understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity… i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Intellectual Ventures v. Symantec, 838 F.3d 1307, 1321; 120 USPQ2d 1353, 1362 (Fed. Cir. 2016)”. Therefore, the claim is not patent eligible.
Regarding Claim 8,
Claim 8 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 8 is directed to a medium, which is directed to a manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
map an input context of an ad opportunity to an output label vector having a plurality of probabilities, each of which predicts a likelihood of a reaction of a corresponding one of the plurality of audiences to the input context of the ad opportunity
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
map an input context of an ad opportunity to an output label vector having a plurality of probabilities, each of which predicts a likelihood of a reaction of a corresponding one of the plurality of audiences to the input context of the ad opportunity (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional elements of “Machine readable and non-transitory medium having information recorded thereon for predictive targeting, wherein the information, when read by the machine, causes the machine to perform the steps”, “initializing a joint predictive model with initial model parameters”, and “the learned joint predictive model is to be used”, as drafted, amount to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Moreover, the claim recites additional element(s) that amount to insignificant extra-solution activities, which do not integrate a judicial exception into a practical application. For example, the additional element of “obtaining training data based on cookieless online traffic, wherein the training data comprises pairs of data, each of the pairs including an ad opportunity context corresponding to an ad served to a corresponding one of a plurality of audiences and a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served, of a corresponding one of the plurality of audiences in the ad opportunity context” amounts to mere data gathering, which is an insignificant extra-solution activity that does not integrate a judicial exception into a practical application. See MPEP 2106.05(g). Further, the additional element of “machine learning, based on the training data, model parameters of the joint predictive model based on the initial model parameters by minimizing a loss in an iterative process” recites an insignificant extra-solution activity which does not integrate a judicial exception into a practical application. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Moreover, the additional element of “obtaining training data based on cookieless online traffic, wherein the training data comprises pairs of data, each of the pairs including an ad opportunity context corresponding to an ad served to a corresponding one of a plurality of audiences and a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served, of a corresponding one of the plurality of audiences in the ad opportunity context” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See MPEP 2106.05(d)(II) “The courts have recognized the following computer functions as well‐understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity… i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Intellectual Ventures v. Symantec, 838 F.3d 1307, 1321; 120 USPQ2d 1353, 1362 (Fed. Cir. 2016)”. Further, the additional element of “machine learning, based on the training data, model parameters of the joint predictive model based on the initial model parameters by minimizing a loss in an iterative process” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See Abraham et al. (US 20230368915 A1 (Filed 2021)) Specification [0070] “Conventional machine learning model training system 100 can train the machine learning model 110 to minimize the (cumulative) loss function 130 by performing multiple iterations of conventional machine learning model training techniques on training data items”. Therefore, the claim is not patent eligible.
Regarding Claim 9,
Claim 9 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 9 is directed to a medium, which is directed to a manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
the reaction includes one of a conversion and a lack of conversion
a label in the label vector indicates whether the reaction is a conversion or a lack of conversion
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
the reaction includes one of a conversion and a lack of conversion (corresponds to evaluation and judgment);
a label in the label vector indicates whether the reaction is a conversion or a lack of conversion (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: See corresponding analysis of claim 8.
Step 2B Analysis: See corresponding analysis of claim 8.
Regarding Claim 10,
Claim 10 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 10 is directed to a medium, which is directed to a manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
an ad opportunity context is characterized based on a plurality of contextual features
each contextual feature is encoded via a first representation of a first dimension
each first representation is characterized by a feature vector of a second dimension, wherein the first dimension is larger than the second dimension
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
an ad opportunity context is characterized based on a plurality of contextual features (corresponds to evaluation and judgment);
each contextual feature is encoded via a first representation of a first dimension (corresponds to evaluation and judgment);
each first representation is characterized by a feature vector of a second dimension, wherein the first dimension is larger than the second dimension (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: See corresponding analysis of claim 8.
Step 2B Analysis: See corresponding analysis of claim 8.
Regarding Claim 11,
Claim 11 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 11 is directed to a medium, which is directed to a manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
predict an output label vector
representing each contextual feature of an ad opportunity context via the first representation weighted by a first coefficient
[representing] each first representation via the feature vector
incorporating interactions between feature vectors of different contextual features of each ad opportunity context weighted by a second set of coefficients
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
predict an output label vector (corresponds to evaluation and judgment);
representing each contextual feature of an ad opportunity context via the first representation weighted by a first coefficient (corresponds to evaluation and judgment);
[representing] each first representation via the feature vector (corresponds to evaluation and judgment);
incorporating interactions between feature vectors of different contextual features of each ad opportunity context weighted by a second set of coefficients (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “the joint predictive model is constructed”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Therefore, the claim is not patent eligible.
Regarding Claim 12,
Claim 12 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 12 is directed to a medium, which is directed to a manufacture, one of the statutory categories.
Step 2A Prong One Analysis: See corresponding analysis of claim 11.
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “initializing values of feature vectors related to the plurality of contextual features”, “initializing values of the first coefficients used to weigh the corresponding plurality of contextual features”, and “initializing values of the second set of coefficients used to weigh the interactions”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Therefore, the claim is not patent eligible.
Regarding Claim 13,
Claim 13 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 13 is directed to a medium, which is directed to a manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
predicting an output label vector with a plurality of probabilities based on current model parameters of the joint predictive model, wherein each probability of the output label vector indicates a likelihood of a reaction from a corresponding one of the plurality of audiences
computing a loss based on the predicted output label vector and the label vector from the pair of data from the training data
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) and/or mathematical concepts (mathematical relationships, mathematical formulas or equations, mathematical calculations) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
predicting an output label vector with a plurality of probabilities based on current model parameters of the joint predictive model, wherein each probability of the output label vector indicates a likelihood of a reaction from a corresponding one of the plurality of audiences (corresponds to evaluation and judgment);
computing a loss based on the predicted output label vector and the label vector from the pair of data from the training data (corresponds to mathematical calculations).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “the steps of obtaining, predicting, computing, and adjusting are repeated in the iterative process until the loss satisfies a pre-determined criterion to generate the joint predictive model with converged model parameters”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Further, the additional element of “adjusting the current model parameters of the joint predictive model by minimizing the loss” recites an insignificant extra-solution activity which does not integrate a judicial exception into a practical application. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Further, the additional element of “adjusting the current model parameters of the joint predictive model by minimizing the loss” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See Abraham et al. (US 20230368915 A1 (Filed 2021)) Specification [0070] “Conventional machine learning model training system 100 can train the machine learning model 110 to minimize the (cumulative) loss function 130 by performing multiple iterations of conventional machine learning model training techniques on training data items”. Therefore, the claim is not patent eligible.
Regarding Claim 14,
Claim 14 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 14 is directed to a medium, which is directed to a manufacture, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
creating first representations for corresponding contextual features of the input context of the ad opportunity
generating an output label vector with respect to the plurality of audiences, based on the joint predictive model with converged model parameters, to predict probabilities of reactions of the respective plurality of audiences to the ad opportunity context
select one or more of the plurality of audiences based on the predicted probabilities in the output label vector
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
creating first representations for corresponding contextual features of the input context of the ad opportunity (corresponds to evaluation and judgment);
generating an output label vector with respect to the plurality of audiences, based on the joint predictive model with converged model parameters, to predict probabilities of reactions of the respective plurality of audiences to the ad opportunity context (corresponds to evaluation and judgment);
select one or more of the plurality of audiences based on the predicted probabilities in the output label vector (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “to enable the DSP to”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Moreover, the claim recites additional element(s) that amount to insignificant extra-solution activities, which do not integrate a judicial exception into a practical application. For example, the additional element of “receiving, from a demand side platform (DSP), an input context of an ad opportunity” and “transmitting the output label vector to the DSP” amounts to mere data gathering, which is an insignificant extra-solution activity that does not integrate a judicial exception into a practical application. See MPEP 2106.05(g). Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Moreover, the additional element of “obtaining training data comprising pairs of data, each of the pairs includes an ad opportunity context corresponding to an ad served to a plurality of audiences and a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served, of a corresponding one of the plurality of audiences in the ad opportunity context” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See MPEP 2106.05(d)(II) “The courts have recognized the following computer functions as well‐understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity… i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Intellectual Ventures v. Symantec, 838 F.3d 1307, 1321; 120 USPQ2d 1353, 1362 (Fed. Cir. 2016)”. Therefore, the claim is not patent eligible.
Regarding Claim 15,
Claim 15 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 15 is directed to a system, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
map an input context of an ad opportunity to an output label vector having a plurality of probabilities, each of which predicts a likelihood of a reaction of a corresponding one of the plurality of audiences to the input context of the ad opportunity
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
map an input context of an ad opportunity to an output label vector having a plurality of probabilities, each of which predicts a likelihood of a reaction of a corresponding one of the plurality of audiences to the input context of the ad opportunity (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional elements of “memory storing computer program instructions”, “one or more processors that, in response to executing the computer program instructions, effectuate operations”, “the learned joint predictive model is to be used”, “machine learning, based on the training data, model parameters of the joint predictive model based on the initial model parameters by minimizing a loss in an iterative process”, and “initializing a joint predictive model with initial model parameters”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Moreover, the claim recites additional element(s) that amount to insignificant extra-solution activities, which do not integrate a judicial exception into a practical application. For example, the additional element of “generating training data based on cookieless online traffic, wherein the training data comprises pairs of data, each of the pairs including an ad opportunity context corresponding to an ad served to a corresponding one of a plurality of audiences and a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served, of a corresponding one of the plurality of audiences in the ad opportunity context” amounts to mere data gathering, which is an insignificant extra-solution activity that does not integrate a judicial exception into a practical application. See MPEP 2106.05(g). Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Moreover, the additional element of “generating training data based on cookieless online traffic, wherein the training data comprises pairs of data, each of the pairs including an ad opportunity context corresponding to an ad served to a corresponding one of a plurality of audiences and a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served, of a corresponding one of the plurality of audiences in the ad opportunity context” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See MPEP 2106.05(d)(II) “The courts have recognized the following computer functions as well‐understood, routine, and conventional functions when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity… i. Receiving or transmitting data over a network, e.g., using the Internet to gather data, Intellectual Ventures v. Symantec, 838 F.3d 1307, 1321; 120 USPQ2d 1353, 1362 (Fed. Cir. 2016)”. Further, the additional element of “machine learning, based on the training data, model parameters of the joint predictive model based on the initial model parameters by minimizing a loss in an iterative process” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See Abraham et al. (US 20230368915 A1 (Filed 2021)) Specification [0070] “Conventional machine learning model training system 100 can train the machine learning model 110 to minimize the (cumulative) loss function 130 by performing multiple iterations of conventional machine learning model training techniques on training data items”. Therefore, the claim is not patent eligible.
Regarding Claim 16,
Claim 16 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 16 is directed to a system, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
the reaction includes one of a conversion and a lack of conversion
a label in the label vector indicates whether the reaction is a conversion or a lack of conversion
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
the reaction includes one of a conversion and a lack of conversion (corresponds to evaluation and judgment);
a label in the label vector indicates whether the reaction is a conversion or a lack of conversion (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: See corresponding analysis of claim 15.
Step 2B Analysis: See corresponding analysis of claim 15.
Regarding Claim 17,
Claim 17 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 17 is directed to a system, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
an ad opportunity context is characterized based on a plurality of contextual features
each contextual feature is encoded via a first representation of a first dimension
each first representation is characterized by a feature vector of a second dimension, wherein the first dimension is larger than the second dimension
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
an ad opportunity context is characterized based on a plurality of contextual features (corresponds to evaluation and judgment);
each contextual feature is encoded via a first representation of a first dimension (corresponds to evaluation and judgment);
each first representation is characterized by a feature vector of a second dimension, wherein the first dimension is larger than the second dimension (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: See corresponding analysis of claim 15.
Step 2B Analysis: See corresponding analysis of claim 15.
Regarding Claim 18,
Claim 18 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 18 is directed to a system, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
predict an output label vector
representing each contextual feature of an ad opportunity context via the first representation weighted by a first coefficient
[representing] each first representation via the feature vector
incorporating interactions between feature vectors of different contextual features of each ad opportunity context weighted by a second set of coefficients
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
predict an output label vector (corresponds to evaluation and judgment);
representing each contextual feature of an ad opportunity context via the first representation weighted by a first coefficient (corresponds to evaluation and judgment);
[representing] each first representation via the feature vector (corresponds to evaluation and judgment);
incorporating interactions between feature vectors of different contextual features of each ad opportunity context weighted by a second set of coefficients (corresponds to evaluation and judgment).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “the joint predictive model is constructed”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Therefore, the claim is not patent eligible.
Regarding Claim 19,
Claim 19 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 19 is directed to a system, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: See corresponding analysis of claim 18.
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “initializing values of feature vectors related to the plurality of contextual features”, “initializing values of the first coefficients used to weigh the corresponding plurality of contextual features”, and “[initializing] values of the second set of coefficients used to weigh the interactions”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Therefore, the claim is not patent eligible.
Regarding Claim 20,
Claim 20 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 20 is directed to a system, which is directed to a machine, one of the statutory categories.
Step 2A Prong One Analysis: The following limitation:
predicting an output label vector with a plurality of probabilities based on current model parameters of the joint predictive model, wherein each probability of the output label vector indicates a likelihood of a reaction from a corresponding one of the plurality of audiences
computing a loss based on the predicted output label vector and the label vector from the pair of data from the training data
as drafted, under the broadest reasonable interpretation, covers mental processes – concepts performed in the human mind (including an observation, evaluation, judgment, opinion) and/or mathematical concepts (mathematical relationships, mathematical formulas or equations, mathematical calculations) but for the recitation of mere instructions to apply an exception language and insignificant extra-solution activity language. In particular, the above limitation in the context of this claim encompasses:
predicting an output label vector with a plurality of probabilities based on current model parameters of the joint predictive model, wherein each probability of the output label vector indicates a likelihood of a reaction from a corresponding one of the plurality of audiences (corresponds to evaluation and judgment);
computing a loss based on the predicted output label vector and the label vector from the pair of data from the training data (corresponds to mathematical calculations).
Step 2A Prong Two Analysis: This judicial exception is not integrated into a practical application. In particular, the claim recites additional elements that amount to recitation of the words “apply it” (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer, which do not integrate a judicial exception into a practical application. See MPEP 2106.05(f). For example, the additional element of “the steps of obtaining, predicting, computing, and adjusting are repeated in the iterative process until the loss satisfies a pre-determined criterion to generate the joint predictive model with converged model parameters”, as drafted, amounts to mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea. Further, the additional element of “adjusting the current model parameters of the joint predictive model by minimizing the loss” recites an insignificant extra-solution activity which does not integrate a judicial exception into a practical application. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. See MPEP 2106.04(d). Even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception.
Step 2B Analysis: As discussed above with respect to integration of the abstract idea into a practical application, the claim recites additional elements that amount to recitation of the words "apply it" (or an equivalent) or are more than mere instructions to implement an abstract idea or other exception on a computer. This has been re-evaluated under step 2B and does not amount to significantly more. See MPEP 2106.05(f). Mere instructions to apply an exception cannot provide an inventive concept. Further, the additional element of “adjusting the current model parameters of the joint predictive model by minimizing the loss” amounts to an insignificant extra-solution activity that is well-understood, routine, and conventional. Similarly, this has been re-evaluated under step 2B and do not amount to significantly more. See Abraham et al. (US 20230368915 A1 (Filed 2021)) Specification [0070] “Conventional machine learning model training system 100 can train the machine learning model 110 to minimize the (cumulative) loss function 130 by performing multiple iterations of conventional machine learning model training techniques on training data items”. Therefore, the claim is not patent eligible.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Ordentlich et al. (US 20190251428 A1 (Published 2019); hereinafter Ordentlich) in view of Allison (US 20110082824 A1), further in view of Gharibshah et al. (“User Response Prediction in Online Advertising” (Published 2021); hereinafter Gharibshah).
Regarding claim 1,
Ordentlich teaches a method implemented on at least one processor, a memory, and a communication platform for predictive targeting (Ordentlich Specification [0007] “a method, implemented on a machine having at least one processor, storage, and a communication platform capable of connecting to a network for obtaining a model”; discloses a method implemented with a processor, storage, and communication platform for advertising (corresponds to a method implemented on at least one processor, a memory, and a communication platform for predictive targeting)), comprising:
obtaining training data comprising pairs of data, each of the pairs includes an ad opportunity context corresponding to an ad served to a … audience … and a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served, of a corresponding one of the plurality of audiences in the ad opportunity context (Ordentlich Specification [0054] “The positive training data set 520 includes pairs of words representing user events that appear close to one another in the temporally ordered session data. For example, the positive training data may include a word corresponding to a query and a word corresponding to a clicked advertisement”; [0049] “The session data also record events that occurred in the session. For each query, user interactions with either the content links or the advertisement are recorded, e.g., in the order of the sequence of events as they occurred and optionally with other useful meta information. Such interactions include clicking on a content link or an advertisement, dwelling on certain content/advertisement, etc”; [0052] “The model training engine 410 receives, at 440, session data as training data and learns from the session data to generate, at 450, the query/ads model 420”; [0046] “The advertisements stored on the advertisement server 240 may include some textual information, e.g., a description of what the advertisement is about as well as additional information such as target audience of the advertisement”; discloses using a training data set including pairs of words (corresponds to obtaining training data comprising pairs of data) representing user queries (corresponds to each of the pairs includes an ad opportunity context corresponding to an ad served to a … audience) and clicked ads/links which would indicate a reaction of a user, e.g., a user having clicking said ad/link (corresponds to a label vector having a plurality of labels, each of which indicates a reaction, with respect to the ad served). Further, advertisements include data for a target audience but not specifically to several target audiences (a corresponding one of the plurality of audiences in the ad opportunity context));
initializing a joint predictive model with initial model parameters (Ordentlich Specification [0062] “Vectors (u vectors and/or v vectors) to be optimized for words from positive/negative examples are first initialized”; [0063] “Based on the vectors generated for ads/links, queries, and subwords, the query/ads model optimization engine 740 learns, via an iterative process, these vectors/parameters”; discloses initializing vectors/parameters for training a machine learning model (corresponds to initializing a joint predictive model with initial model parameters)); and
machine learning, based on the training data, model parameters of the joint predictive model based on the initial model parameters by minimizing a loss in an iterative process (Ordentlich Specification [0061] “the learning based training engine 540 comprises a positive example vector generator/updater 710, a negative example vector generater/updater 730, a subword vector generator/updater 720, a subword vector combiner 760, and a query/ads model optimization engine 740”; [0062] “the learning based training engine 540 takes the positive and negative training data sets 520 and 530 as input and learns various parameters of the query/ads model 420 so that the model parameters are trained so that they maximize an objective function or minimize a loss function (discussed below) based on positive and negative examples from the training data”; [0063] “Based on the vectors generated for ads/links, queries, and subwords, the query/ads model optimization engine 740 learns, via an iterative process”; discloses training a machine learning model using positive and negative training data sets (corresponds to machine learning, based on the training data, model parameters of the joint predictive model based on the initial model parameters) with the goal of minimizing a loss function in an iterative learning process (corresponds to minimizing a loss in an iterative process)), wherein
the learned joint predictive model is to be used to map an input context of an ad opportunity to an output label vector having a plurality of probabilities, each of which predicts a likelihood of a reaction of a corresponding one of the plurality of audiences to the input context of the ad opportunity (Ordentlich Specification [0053] “With respect to a query received at 460, the query/ads model based ad selector 430 receives, at 470, a set of candidate ads. For the query and the candidate ads, corresponding vectors for subwords of the query, the query, and the candidate ads are obtained at 480 based on the query/ads model 420. Based on the obtained vectors, the query/ads model based ad selector 430 selects, at 490, a most relevant ad”; [0077] “the objective is to maximize a modeled probability of co-occurrence of a “context” word (e.g., clicked ad) and a central word (e.g., query) in a positive example pair”; [0101] “With the vectors for subwords of the input query, the vector(s) for the query, and the vectors for each of the candidate ads, the model based query/ad affinity estimator 1460 computes, at 1570, an affinity score for each pair of the input query and one of the candidate ads. Based on such affinity scores for all the candidate ads, the matching ad selector 1470 may then select, at 1580, one or more matching advertisements”; [0046] “The advertisements stored on the advertisement server 240 may include some textual information, e.g., a description of what the advertisement is about as well as additional information such as target audience of the advertisement”; discloses a machine learning model used to match queries with candidate ads which have a computed affinity for the input query (corresponds to input context of an ad opportunity) and the candidate ad which are stored in a vector indicating a probability of the co-occurrence of a clicked ad (corresponds to a reaction) and an input query (corresponds to the learned joint predictive model is to be used to map an input context of an ad opportunity to an output label vector having a plurality of probabilities). Positive pairs of clicked ads and queries would have a maximized probability to indicate a target audience having clicked on a particular ad given a user’s input query (corresponds to a plurality of probabilities, each of which predicts a likelihood of a reaction of a corresponding one of the plurality of audiences to the input context of the ad opportunity; for more please see Ordentlich Specification [0077])).
Ordentlich does not teach wherein the draining data is based on a cookieless online traffic.
Allison teaches wherein training data is based on a cookieless online traffic (¶62) for improved user profiling (¶62).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to utilize cookieless online traffic in the training data of Ordentlich in order to improve user profiling.
Ordentlich appears to not explicitly disclose an ad served to corresponding one of a plurality of audiences.
However, Gharibshah teaches an ad served to a corresponding one of a plurality of audiences (Gharibshah P.1, Sec.1, Para.2 “the ads to be displayed to audience (i.e., users)”; P.4, Para.2 “using context to find users’ preference plays an essential role for user response prediction. The information from publisher websites is usually obtained from crawling the web-pages to summarize the context. It is then complimented by online analysis of cookie data and browsing history made by users. Such information allows system to identify user interest and response regarding ad impression”; discloses gathering information in order to identify users’ preferences regarding ad impressions which would imply the ad in question is being served to several audiences (corresponds to an ad served to a plurality of audiences)).
Ordentlich and Gharibshah are considered to be analogous to the claimed invention because they are in the same field of utilizing machine learning models with regard to advertising. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Ordentlich to incorporate the teachings of Gharibshah. Doing so could improve the performance of ads disclosed by Ordentlich by gathering context of more audiences that leave a positive impression when being served a particular ad, as suggested by Gharibshah (Gharibshah P.4, Para.2 “serving users with ads best matching to their preference is of interests to both advertisers and publishers”).
Regarding claim 2,
Ordentlich in view of Gharibshah teaches the method of claim 1.
Ordentlich further teaches the reaction includes one of a conversion and a lack of conversion (Ordentlich Specification [0047] “a user may click on an advertisement displayed which may ultimately lead to a click through or conversion, i.e., a purchase made on the product/service advertised. As another example, the dwell time that the user spent on a display advertisement (e.g., detected by computing the length of time during which the cursor dwells on the advertisement) may also indicate that the user is interested in the product or service of the advertisement”; [0058] “the session data 310 provide positive pairings between queries and ads (and associated hyperlinks). In other embodiments, session data 310 may also include session information related to ads that were placed yet not clicked”; discloses collecting session data for advertisements which indicate whether a user clicks on or dwells or does not click on said advertisement (corresponds to the reaction includes one of a conversion and a lack of conversion)); and
a label in the label vector indicates whether the reaction is a conversion or a lack of conversion (Ordentlich Specification [0049] “user interactions with either the content links or the advertisement are recorded, e.g., in the order of the sequence of events as they occurred and optionally with other useful meta information. Such interactions include clicking on a content link or an advertisement, dwelling on certain content/advertisement”; [0051] “vectors for the query as a whole in matching the query with vectors for advertisements”; [0052] “the query/ads model 420 involves various parameters and such parameters are optimized via learning (from the session data 310). Such parameters include vectors for subwords of queries, vectors for words from advertisements”; discloses vectors containing advertisement information which, in the process of machine learning, are optimized with respect to session data which includes whether an advertisement was clicked on (conversion) or dwelled on and not clicked (lack of conversion) (correspond to a label in the label vector indicates whether the reaction is a conversion or a lack of conversion)).
Regarding claim 3,
Ordentlich in view of Gharibshah teaches the method of claim 1.
Ordentlich further teaches an ad opportunity context is characterized based on a plurality of contextual features, wherein
each contextual feature is encoded via a first representation of a first dimension, and
each first representation is characterized by a feature vector of a second dimension … (Ordentlich Specification [0062] “Vectors of queries/ads/links and subwords may have some specified dimensions, e.g., 300 or 500, and the values in each of such vectors in each dimension are to be learned via training”; [0079] “an English word or unigram may be represented by a vector with a dimension of 200. For a string such as a phrase or a hyperlink, a vector may have a higher dimension, e.g., 300”; discloses vectors being used to represent queries, advertisements, and links (corresponds to an ad opportunity context is characterized based on a plurality of contextual features). Contextual features such as queries, advertisements, and links must be in one dimension (corresponds to each contextual feature is encoded via a first representation of a first dimension) before being represented in vectors of a specified dimension (corresponds to each first representation is characterized by a feature vector of a second dimension)).
Ordentlich appears to not explicitly disclose the first dimension is larger than the second dimension.
However, Gharibshah teaches the first dimension is larger than the second dimension (Gharibshah P.8, Para.1 “the common approach for many classification based methods is employing the embedding step to generate condensed embedding vectors”; discloses condensing embedding vectors for prediction models. In this case a vector has a first representation with a larger dimension and is condensed to an embedding vector of a smaller dimension (corresponds to the first dimension is larger than the second dimension; for examples of condensed embedding vectors, please see Gharibshah, P.8, Fig.3)).
Ordentlich and Gharibshah are considered to be analogous to the claimed invention because they are in the same field of utilizing machine learning models with regard to advertising. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Ordentlich to incorporate the teachings of Gharibshah. Doing so could improve the performance of models disclosed by Ordentlich by condensing the information stored in model parameters to avoid over-fitting, as suggested by Gharibshah (Gharibshah P.9, Sec.3.2.2, Para.1 “The combination of different features in modeling lead to various compound embedding layer for input data to generate a condensed feature representation, with pooling being employed to reduce parameters and cope with over-fitting”).
Regarding claim 4,
Ordentlich in view of Gharibshah teaches the method of claim 3.
Ordentlich further teaches the joint predictive model is constructed to predict an output label vector (Ordentlich Specification [0101] “the model based query/ad affinity estimator 1460 computes, at 1570, an affinity score for each pair of the input query and one of the candidate ads”; [0061] “positive and negative training data sets are then used for training the query/ad model 420, which is learned from the training data by maximizing the modeled probabilities for the positive examples from the positive training data set and minimizing the modeled probabilities for the negative examples”; discloses a neural network model which outputs probabilistic scores such ass affinity scores or modeled probability distributions (corresponds to the joint predictive model is constructed to predict an output label vector)) by
representing each contextual feature of an ad opportunity context via the first representation weighted by a first coefficient (Ordentlich Specification [0062] “Parameters to be optimized via learning may be specified in a storage or archive 750 (optimization parameters). In some embodiments, such optimization parameters include vectors for queries/ads/links, vectors for subwords of the queries”; [0073] “the combining function m may correspond to more complex, e.g., a component-wise weighted sum of u vectors for subwords”; discloses optimization parameters which include vectors for queries, ads, and links (corresponds to representing each contextual feature of an ad opportunity context via the first representation). Further discloses uses weighted sums of vectors (corresponds to weighted by a first coefficient)), and
each first representation via the feature vector (Ordentlich Specification [0062] “Vectors (u vectors and/or v vectors) to be optimized for words from positive/negative examples are first initialized at 820”; [0051] “vectors for the query as a whole in matching the query with vectors for advertisements”; discloses u and v vectors for representing queries and ads (corresponds to representing … each first representation via the feature vector)); and
incorporating interactions between feature vectors of different contextual features of each ad opportunity context weighted by a second set of coefficients (Ordentlich Specification [0051] “a vector for the query that gives rise to the subwords may be obtained by combining the subword vectors for the subwords, i.e., u(Q)=combiner (u(m.sub.i), i=1, 2, . . . , k), where Q denotes a query”; [0073] “the combining function m may correspond to more complex, e.g., a component-wise weighted sum of u vectors for subwords”; [0094] “Various parameters associated with the CNN model may be trained, e.g., the coefficients of the filters and of the fully connected layers as well as the unigrams vectors”; [0093] “Each of the coefficients in each of the filters may also be optimized during the training”; discloses processes for weighting vectors including weighting u and v vectors for subwords and advertisements (incorporating interactions between feature vectors of different contextual features of each ad opportunity context) by utilizing optimized coefficients (corresponds to weighted by a second set of coefficients)).
Regarding claim 5,
Ordentlich in view of Gharibshah teaches the method of claim 4.
Ordentlich further teaches the step of initializing comprises:
initializing values of feature vectors related to the plurality of contextual features (Ordentlich Specification [0062] “Vectors (u vectors and/or v vectors) to be optimized for words from positive/negative examples are first initialized at 820”; discloses initialization of u and v feature vectors for subwords and advertisements (corresponds to initializing values of feature vectors related to the plurality of contextual features)); and
initializing values of the first coefficients used to weigh the corresponding plurality of contextual features (Ordentlich Specification [0073] “the weights used in a combiner that uses component-wise weighted sum may be optimized during training”; discloses optimizing weights used with u and v feature vectors which requires initialization (corresponds to initializing values of the first coefficients used to weigh the corresponding plurality of contextual features)); and
initializing values of the second set of coefficients used to weigh the interactions (Ordentlich Specification [0066] “both u and v vectors for ads/links, v vectors for queries, u vectors for subwords of queries, as well as u vectors for queries (derived by combining u vectors of the subwords of the queries) … the ad/link/query vector initializer 910 initializes these vectors according to an ad/link/query vector configuration 930, which may specify the dimension of the vectors and may also provide some initial seeds or default values for the attributes of the vectors”; [0093] “Each of the coefficients in each of the filters may also be optimized during the training”; discloses utilizing optimized filter coefficients for weighting u and v vectors which requires initialization (corresponds to initializing values of the second set of coefficients used to weigh the interactions)).
Regarding claim 6,
Ordentlich in view of Gharibshah teaches the method of claim 1.
Ordentlich further teaches wherein the step of machine learning comprises:
obtaining an ad opportunity context from a pair of data in the training data (Ordentlich Specification [0057] “the positive training data set comprises pairs, each of which includes a query or queries ... clicked ad(s) and clicked content link(s)”; [0061] “Such generated positive and negative training data sets are then used for training”; discloses a training data set comprising pairs (corresponds to a pair of data in the training set) which includes queries indicated clicked ads and content (corresponds to obtaining an ad opportunity context));
predicting an output label vector with a plurality of probabilities based on current model parameters of the joint predictive model, wherein each probability of the output label vector indicates a likelihood of a reaction from a corresponding one of the plurality of audiences (Ordentlich Specification [0063] “during each iteration of learning at 860 during the optimization process, an objective function … may be assessed in each iteration based on current vectors and parameters against some convergence conditions. If the conditions are not met (i.e., no convergence), determined at 870, the query/ads model optimization engine 740 proceeds to modify or adjust, at 880, the vectors/parameters in a manner to drive the training towards convergence”; [0076] “the w.sub.i,j and w.sub.i,k correspond to a positive pair (e.g., a query and a clicked ad/link) from the positive training data set 520 and w.sub.i,j and {tilde over (w)} is a negative pair from the negative training data set 530, with {tilde over (w)} being selected randomly from the vocabulary according to a probability distribution … The training may be carried out using some known algorithms such as minibatch SGD optimization with respect to the various word and subword vectors”; [0078] “the above optimization scheme tends to maximize the modeled probability with respect to the positive examples … and at the same time tends to minimize the probability with respect to the negative examples”; discloses an iterative training process which models probability distributions of word and subword vectors (corresponds to predicting an output label vector with a plurality of probabilities based on current model parameters of the joint predictive model) in order to indicate probability of an ad being clicked (corresponds to each probability of the output label vector indicates a likelihood of a reaction) with respect to information indicating a target audience (corresponds to from a corresponding one of the plurality of audiences; for more please see Ordentlich Specification [0046]));
computing a loss based on the predicted output label vector and the label vector from the pair of data from the training data (Ordentlich Specification [0078] “the gradient descent approach is used to train a logistic regression classifier … The objective (2) corresponds to optimizing the log-loss of such a classifier”; [0062] “the learning based training engine 540 takes the positive and negative training data sets 520 and 530 as input and learns various parameters of the query/ads model 420 so that the model parameters are trained so that they maximize an objective function or minimize a loss function”; discloses a training process which takes pairs of data in positive and negative training sets of vectors in order to calculate and minimize a loss function (corresponds to computing a loss based on the predicted output label vector and the label vector from the pair of data from the training data));
adjusting the current model parameters of the joint predictive model by minimizing the loss (Ordentlich Specification [0062] “the learning based training engine 540 takes the positive and negative training data sets 520 and 530 as input and learns various parameters of the query/ads model 420 so that the model parameters are trained so that they maximize an objective function or minimize a loss function”; [0063] “the query/ads model optimization engine 740 proceeds to modify or adjust, at 880, the vectors/parameters in a manner to drive the training towards convergence”; discloses a training process which learns and adjusts model parameters in order to minimize a loss function (corresponds to adjusting the current model parameters of the joint predictive model by minimizing the loss)), wherein
the steps of obtaining, predicting, computing, and adjusting are repeated in the iterative process until the loss satisfies a pre-determined criterion to generate the joint predictive model with converged model parameters (Ordentlich Specification [0063] “the query/ads model optimization engine 740 learns, via an iterative process, these vectors/parameters … during each iteration of learning at 860 during the optimization process, an objective function (discussed below in detail) may be assessed in each iteration based on current vectors and parameters against some convergence conditions. If the conditions are not met (i.e., no convergence), determined at 870, the query/ads model optimization engine 740 proceeds to modify or adjust, at 880, the vectors/parameters in a manner to drive the training towards convergence”; [0076] “The training may be carried out using some known algorithms such as minibatch SGD optimization”; Ordentlich discloses the steps of obtaining, predicting, computing, and adjusting as discussed above (corresponds to the steps of obtaining, predicting, computing, and adjusting). Ordentlich’s training process is performed iteratively (corresponds to repeated in the iterative process) such that in each iteration, the model is assessed as to whether convergence has occurred based on an objective function or loss function (corresponds to until the loss satisfies a pre-determined criterion to generate the joint predictive model with converged model parameters; for more please see Ordentlich Specification [0063])).
Regarding claim 7,
Ordentlich in view of Gharibshah teaches the method of claim 1.
Gharibshah further teaches receiving, from a demand side platform (DSP), an input context of an ad opportunity (Gharibshah P.40, Sec.B.3.2, Para.1 “An ad exchange casts auction for bid request triggered by SSPs to DSPs to select the display of ad on the publisher’s website”; discloses a demand side platform which is given a bid request to select an ad to display (corresponds to receiving, from a demand side platform (DSP), an input context of an ad opportunity));
creating first representations for corresponding contextual features of the input context of the ad opportunity (Gharibshah P.19, Fig.8, “The structure of embedding layer to generate dense embedding vectors. It includes a linear mapping from discrete categorical features represented by one-hot-embedded vectors to dense numerical embedding vectors.”; P.40, Sec.B.3, Para.1 “The source of features for user response prediction task is related to information transferred in online advertising ecosystem”; discloses an embedding layer for generating embedding vectors for categorical features relating to online advertising (creating first representations for corresponding contextual features of the input context of the ad opportunity));
generating an output label vector with respect to the plurality of audiences, based on the joint predictive model with converged model parameters, to predict probabilities of reactions of the respective plurality of audiences to the ad opportunity context (Gharibshah P.14, Para.2 “The predicted probability of 𝑥𝑖 belonging to class 1 is modeled by Sigmoid function as”; (Equation 8; reproduced below);
PNG
media_image1.png
114
589
media_image1.png
Greyscale
P.7, Para.2 “For the prediction task, it will output probability of users making an interaction (e.g. a click) on items in the list [of promoted/recommended products]”; P.7, Fig.2 “The output can be considered as two types of user responses a) a scalar value of predicted score for an interaction between given user 𝑢𝑖 and item 𝐼𝑗”; discloses modeling predicted probabilities of users interacting with advertised products using a defined sigmoid function (corresponds to generating an output label vector with respect to the plurality of audiences) using a trained logistic regression model (corresponds to the joint predictive model with converged model parameters) which performs a prediction task to model probabilities of users interacting with recommended products (corresponds to predict probabilities of reactions of the respective plurality of audiences to the ad opportunity context));
transmitting the output label vector to the DSP to enable the DSP to select one or more of the plurality of audiences based on the predicted probabilities in the output label vector (Gharibshah P.40, Sec.B.3.1 “information regarding visiting user and ad slots are transferred to relevant DSP nodes through ad exchange network … user information such as device types, user agent, browser information etc”; P.7, Para.2 “For the prediction task, it will output probability of users making an interaction (e.g. a click) on items in the list [of promoted/recommended products]” Sec.B.3.2, Para.1 “An ad exchange casts auction for bid request triggered by SSPs to DSPs to select the display of ad on the publisher’s website”; discloses transferring information regarding user information and advertisements to a demand side platform (corresponds to transmitting the output label vector to the DSP) to enable the DSP to select which ads and thus which audiences/users to target using predicted probabilities of users making an interaction with an advertisement (corresponds to enable the DSP to select one or more of the plurality of audiences based on the predicted probabilities in the output label vector)).
Ordentlich and Gharibshah are considered to be analogous to the claimed invention because they are in the same field of utilizing machine learning models with regard to advertising. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Ordentlich to incorporate the teachings of Gharibshah. Doing so could extend the techniques of selective advertising disclosed by Ordentlich to demand side platforms which utilize predictions made by neural networks for ad selection, as suggested by Gharibshah (Gharibshah P.7, Para.1 “the predicted probability is not only used as an indicator to present user preferences, it is also involved in bidding strategies to determine the revenue of advertiser and publishers”).
Regarding claims 8-14,
Claims 8-14 are rejected under the same grounds as claims 1-7 respectively.
Per claim 8, Ordentlich teaches machine readable and non-transitory medium having information recorded thereon for predictive targeting, wherein the information, when read by the machine, causes the machine to perform the steps (Ordentlich Specification [0010] “a machine-readable, non-transitory and tangible medium having data recorded thereon for obtaining a model for identifying content matching a query, wherein the medium, when read by the machine, causes the machine to perform a series of steps. Training data are received which include queries, advertisements, and hyperlinks”; discloses a machine-readable, non-transitory medium storing data and performing steps for advertising (corresponds to machine readable and non-transitory medium having information recorded thereon for predictive targeting, wherein the information, when read by the machine, causes the machine to perform the steps)).
Regarding claims 15-20,
Claims 15-20 is rejected on the same grounds as claims 1-6 respectively.
Per claim 15, Ordentlich teaches a system for predictive targeting, comprising:
memory storing computer program instructions and one or more processors that, in response to executing the computer program instructions, effectuate operations (Ordentlich Specification [0007] “a method, implemented on a machine having at least one processor, storage, and a communication platform capable of connecting to a network for obtaining a model”)
Response to Arguments
Applicant's arguments filed 03/09/2026 have been fully considered but they are not persuasive.
Applicant argues that “big data analytics and modeling thereof cannot be performed in the human mind or by a human with pencil and paper”. The claim does not recite the usage of data in such a way as to convey that it is limited to amounts of data which are too large or processes too complex to be done in the human mind or by a human with pencil and paper.
Applicant argues that “obtaining training data based on a cookieless online traffic” and “machine learning, based on the training data, model parameters of the joint predictive model based on the initial model parameters by minimizing a loss in an iterative process” cannot be performed in the human mind or by pencil and paper and are therefore not abstract ideas. These are not purported to be abstract ideas herein. Therefore, the argument is moot.
Applicant has argued that the claims provide an improvement to known technical problems associated with a cookieless environment. The claims merely recite the use of cookieless data. The claim must include the components or steps of the invention that provide the improvement described in the specification (see MPEP 2106.05(a)) and the details of the unconventional technical solution must be expressed in the claim or technical improvements realized by the claim over the prior art must be identified (see MPEP 2106.05(a)). The inclusion of cookieless data does not amount to the inclusion of the components leading to the improvement discussed by Applicant.
Applicant has argued that the Office failed to consider Applicant’s claims as an ordered combination and as a whole by incorrectly and improperly identifying that the additional elements do not amount to significantly more than the judicial exception. Per the rejections under 35 USC 101 herein, the claims as a whole are analyzed with respect to step 2B and reevaluated as to whether the additional elements amount to significantly more. Applicant argues that the steps of obtaining training data and machine learning are significantly more because they get around challenges faced in a cookieless environment. As noted above, the inclusion of cookieless data into the claim does not import the entirety of the necessary components purported in the Specification which lead to the improvement.
Applicant has argued that none of Ordentlich or Gharibshah teach the cookieless online traffic. None of Ordentlich or Gharibshah are relied upon to teach this feature and therefore the argument is moot.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SCHYLER S SANKS whose telephone number is (571)272-6125. The examiner can normally be reached 06:30 - 15:30 Central Time, M-F.
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, Michael Huntley can be reached at (303) 297-4307. 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.
/SCHYLER S SANKS/Primary Examiner, Art Unit 2129
Prosecution Insights