DETAILED ACTION
This action is responsive to communications filed on June 5, 2023. This action is made Non-Final.
Claims 1-20 are pending in the case.
Claims 1, 8, and 15 are independent claims.
Claims 1-20 are rejected.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Receipt is acknowledged of papers submitted under 35 U.S.C. 119(a)-(d), which papers have been placed of record in the file.
Information Disclosure Statement
The information disclosure statement (IDS(s)) submitted on 06/27/2023 and 05/15/2024 is/are in compliance with the provisions of 37 C.F.R. 1.97. Accordingly, the IDS(s) is/are being considered by the examiner.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1, 4-7, 8, 11-14, 15, and 17-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Guo et al., 2022/0198289 (“Guo”), and further in view of Walters et al., US Patent 11,042,700 (“Walters”).
Claim 1:
Gou teaches or suggests a method for training a content recommendation model, comprising:
obtaining a sample data set, the sample data set comprising a historical account and a historical recommendation content, and interaction data between the historical account and the historical recommendation content being labeled (see para. 0146 - Obtain a training sample, where the training sample includes a sample user behavior log, position information of a sample recommended object, and a sample label, and the sample label is used to indicate whether a user selects the sample recommended object; para. 0149 - user profile information may also be referred to as a crowd profile, and is a labeled profile abstracted based on information such as demographics, social relationships, preferences, habits, and consumption behaviors of the user; para. 0150 - characteristic information of the recommended object may be a category of the recommended object, or may be an identifier of the recommended object, for example, an ID of a historical recommended object; para. 0152 - one piece of training sample data may include context information (for example, time), position information, user information, and commodity information.);
inputting the sample data set into a probability prediction model to output a probability prediction result, the probability prediction result indicating a predicted probability of the historical account selecting the historical recommendation content (see Fig. 5; para. 0160 - training ... a recommendation model by using the sample user behavior log ... as input data and using the sample label as a target output value, to obtain a trained recommendation model ... the recommendation model is used to predict, when the user pays attention to the target recommended object, a probability that the user selects the target recommended object; para. 0165 - based on a difference between the actual sample label and a jointly predicted selection probability. the model parameters of the position aware model and the recommendation model may be obtained through a plurality of iterations based on the difference between the sample label and the jointly predicted selection probability; para. 0166 - predicted selection probability may be a predicted probability that includes ... that the user selects the sample object.);
input the sample data set ... for which the historical account views the historical recommendation content (see para. 0160 - training on a position aware model and a recommendation model by using the sample user behavior log and the position information of the sample recommended object as input data and using the sample label as a target output value, to obtain a trained recommendation model, where the position aware model is used to predict probabilities that the user pays attention to a target recommended object when the target recommended object is at different positions; para. 0165 – joint training may be training parameters of the position aware model and the actual user recommendation model based on a difference between the actual sample label and a jointly predicted selection probability that include position information.);
determining a probability prediction loss corresponding to the probability prediction result (see para. 0092 - updated based on a difference between the current predicted value and the target value. loss function (loss function) or an objective function (objective function). The loss function and the objective function are important equations used to measure the difference between the predicted value and the target value; para. 0165 - based on a difference between the actual sample label and a jointly predicted selection probability. the model parameters of the position aware model and the recommendation model may be obtained through a plurality of iterations based on the difference between the sample label and the jointly predicted selection probability; para. 0166 - predicted selection probability may be a predicted probability that includes ... that the user selects the sample object; para. 0211 - training may be training model parameters ... the recommendation model based on a difference between the sample label and a jointly predicted selection probability.);
training the probability prediction model based on the probability prediction loss and ... to obtain the content recommendation model, the content recommendation model predicting a recommendation probability of recommending a target content to a target account (see Fig. 2, 5-8; para. 0092 - updated based on a difference between the current predicted value and the target value. loss function (loss function) or an objective function (objective function). The loss function and the objective function are important equations used to measure the difference between the predicted value and the target value; para. 0136 - performing joint training on a ... model and a recommendation model by using the sample user behavior log and ... information of the sample recommended object as input data and using the sample label as a target output value, to obtain a trained recommendation model; para. 0165 - based on a difference between the actual sample label and a jointly predicted selection probability. the model parameters of the position aware model and the recommendation model may be obtained through a plurality of iterations based on the difference between the sample label and the jointly predicted selection probability; para. 0166 - predicted selection probability may be a predicted probability that includes ... that the user selects the sample object; para. 0211 - training may be training model parameters ... the recommendation model based on a difference between the sample label and a jointly predicted selection probability, and the jointly predicted selection probability is obtained based on output data of the ... model and the recommendation model; para. 0229 - input the user characteristic information, the context information, and the candidate recommended object set into a pre-trained recommendation model to obtain a probability that the to-be-processed user selects a candidate recommended object in the candidate recommended object set, where the pre-trained recommendation model is used to predict, when the user pays attention to a target recommended commodity, a probability that the user selects the target recommended object; and obtain a recommendation result of the candidate recommended object based on the probability that the to-be-processed user selects the candidate recommended object, where a model parameter of the pre-trained recommendation model is obtained by performing joint training.).
Guo does not explicitly disclose into a duration prediction model to output a duration prediction result, the duration prediction result indicating predicted duration; and duration prediction loss corresponding to the duration prediction result; and the duration prediction loss.
Walters teaches or suggests into a duration prediction model to output a duration prediction result, the duration prediction result indicating predicted duration; and duration prediction loss corresponding to the duration prediction result; and the duration prediction loss (see col. 11, lines 27-41 - Metrics of user focus may include the amount of time that the text block stays on the screen (e.g., the amount of time being greater than or equal to a threshold amount of time or otherwise satisfying the threshold amount. user-specific focus score may range from 0.0 to 1.0; col. 12, lines 2-43 - compute user-specific text block scores for each text block in a document. The trained neural network may then be used to predict labels such as "of user interest" or "not of user interest" for each text block in a new document based on these viewing times (or other measurements of viewing behavior). Set of user-specific focus scores that ranges between 0.0 (indicating no interest) and 1.0 (indicating high interest) for each of a training set of text blocks may be used for training a neural network. determine an initial set of focus scores and determine a loss function value by taking a difference between the initial set of focus scores and the reference outputs. The system may propagate the loss function value through the neurons of the neural network and re-calculate the neural network parameters using gradient descent; col. 13, lines 23-37 - output of the neural network used may be a label, such as "of interest to user" or "not of interest to user," and the user specific criterion may be based on the label determined by the neural network. output of the neural network may be a quantitative value used to determine whether the text block score satisfies a user-specific display criterion. system may determine that a user-specific display criterion having an associated threshold value of 50% is satisfied if a user-specific text block score determined by the neural network is greater than 50%.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Guo, to include into a duration prediction model to output a duration prediction result, the duration prediction result indicating predicted duration; and duration prediction loss corresponding to the duration prediction result; and the duration prediction loss for the purpose of efficiently determining prediction scores based on view or interaction time period for determining loss values and used in model training, improving model accuracy, as taught by Walters (col. 12).
Claim(s) 8 and 15:
Claim(s) 8 and 15 correspond to Claim 1, and thus, Guo and Walters teach or suggest the limitations of claim(s) 8 and 15 as well.
Claim 4:
Guo further teaches or suggests extracting a semantic feature corresponding to the historical recommendation content, an account attribute feature corresponding to the historical account, and a historical interaction feature corresponding to the historical recommendation content, the semantic feature, the account attribute feature, and the historical interaction feature being used as input features to the probability prediction model (see para. 0146 - Obtain a training sample, where the training sample includes a sample user behavior log, position information of a sample recommended object, and a sample label, and the sample label is used to indicate whether a user selects the sample recommended object; para. 0149 - user profile information may also be referred to as a crowd profile, and is a labeled profile abstracted based on information such as demographics, social relationships, preferences, habits, and consumption behaviors of the user; para. 0150 - characteristic information of the recommended object may be a category of the recommended object, or may be an identifier of the recommended object, for example, an ID of a historical recommended object; para. 0152 - one piece of training sample data may include context information (for example, time), position information, user information, and commodity information; para. 0160 - Perform joint training on a position aware model and a recommendation model by using the sample user behavior log and the position information of the sample recommended object as input data and using the sample label as a target output value, to obtain a trained recommendation model; para. 0161 - the probability that the user selects the target recommendation may be a probability that the user clicks the target object, for example, may be a probability that the user downloads the target object, or a probability that the user browses the target object. Alternatively, the probability that the user selects the target object may be a probability that the user performs a user operation on the target object; para. 0198 – recommendation system constructs an input vector that is based on common characteristics such as a user characteristic, a commodity characteristic, and context information; para. 0204 - characteristic information of the candidate recommended object may be a category of the candidate recommended object, or may be an identifier of the candidate recommended object, for example, an ID of a commodity; para. 0205 - a labeled profile abstracted based on information such as demographics, social relationships, preferences, habits, and consumption behaviors of the user.).
Walters further teaches further teaches or suggests and the duration prediction model (see col. 11, lines 27-41 - Metrics of user focus may include the amount of time that the text block stays on the screen (e.g., the amount of time being greater than or equal to a threshold amount of time or otherwise satisfying the threshold amount. user-specific focus score may range from 0.0 to 1.0; col. 12, lines 2-43 - compute user-specific text block scores for each text block in a document. The trained neural network may then be used to predict labels such as "of user interest" or "not of user interest" for each text block in a new document based on these viewing times (or other measurements of viewing behavior). Set of user-specific focus scores that ranges between 0.0 (indicating no interest) and 1.0 (indicating high interest) for each of a training set of text blocks may be used for training a neural network. determine an initial set of focus scores and determine a loss function value by taking a difference between the initial set of focus scores and the reference outputs. The system may propagate the loss function value through the neurons of the neural network and re-calculate the neural network parameters using gradient descent; col. 13, lines 23-37 - output of the neural network used may be a label, such as "of interest to user" or "not of interest to user," and the user specific criterion may be based on the label determined by the neural network. output of the neural network may be a quantitative value used to determine whether the text block score satisfies a user-specific display criterion. system may determine that a user-specific display criterion having an associated threshold value of 50% is satisfied if a user-specific text block score determined by the neural network is greater than 50%.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Guo, to include and the duration prediction model for the purpose of efficiently determining prediction scores based on view or interaction time period for determining loss values and used in model training, improving model accuracy, as taught by Walters (col. 12).
Claim(s) 11 and 17:
Claim(s) 11 and 17 correspond to Claim 4, and thus, Guo and Walters teach or suggest the limitations of claim(s) 11 and 17 as well.
Claim 5:
Guo further teaches or suggests wherein the training the probability prediction model based on the prediction loss to obtain the content recommendation model comprises: performing, based on the prediction loss, gradient adjustment on model parameters of the probability prediction model to obtain the content recommendation model (see para. 0092 - updated based on a difference between the current predicted value and the target value. loss function (loss function) or an objective function (objective function). The loss function and the objective function are important equations used to measure the difference between the predicted value and the target value; para. 0136 - performing joint training on a ... model and a recommendation model by using the sample user behavior log and ... information of the sample recommended object as input data and using the sample label as a target output value, to obtain a trained recommendation model; para. 0165 - based on a difference between the actual sample label and a jointly predicted selection probability. the model parameters of the position aware model and the recommendation model may be obtained through a plurality of iterations based on the difference between the sample label and the jointly predicted selection probability; para. 0166 - predicted selection probability may be a predicted probability that includes ... that the user selects the sample object; para. 0191 - parameters may be updated by using a sampling gradient descent method; para. 0211 - training may be training model parameters ... the recommendation model based on a difference between the sample label and a jointly predicted selection probability, and the jointly predicted selection probability is obtained based on output data of the ... model and the recommendation model.).
Walters also teaches or suggests performing, based on the prediction loss, gradient adjustment on model parameters of the probability prediction model to obtain the content recommendation model (see col. 12, lines 19-43 - loss function value may then be propagated back via a backpropagation operation to update the parameters of the neural network using a gradient descent method. This process may be repeated during training to update the parameters of the neural network until the neural network prediction matches the reference output within an accuracy threshold. determine an initial set of focus scores and determine a loss function value by taking a difference between the initial set of focus scores and the reference outputs. The system may propagate the loss function value through the neurons of the neural network and re-calculate the neural network parameters using gradient descent.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Guo, to include performing, based on the prediction loss, gradient adjustment on model parameters of the probability prediction model to obtain the content recommendation model for the purpose of efficiently determining prediction scores based on view or interaction time period for determining loss values and used in model training, improving model accuracy, as taught by Walters (col. 12).
Claim(s) 12 and 18:
Claim(s) 12 and 18 correspond to Claim 5, and thus, Guo and Walters teach or suggest the limitations of claim(s) 12 and 18 as well.
Claim 6:
Guo further teaches or suggests training, through the prediction loss, the ... prediction model applied to the ith iterative training to obtain an iteratively updated ... prediction model, the iteratively updated ... prediction model being applied to (i+1)th iterative training (see para. 0016 - model parameters of the position aware model and the recommendation model may be obtained through a plurality of iterations based on the difference between the sample label and the jointly predicted selection probability and by using a back propagation algorithm; para. 0075 - module that is in the personalized recommendation system and that iteratively updates a parameter of a recommendation model by using the machine learning algorithm based on the historical data of the user until a specified requirement is met.).
Walters further teaches or suggests duration prediction model ... duration prediction model ... updated duration prediction model (see col. 11, lines 27-41 - Metrics of user focus may include the amount of time that the text block stays on the screen (e.g., the amount of time being greater than or equal to a threshold amount of time or otherwise satisfying the threshold amount. user-specific focus score may range from 0.0 to 1.0; col. 12, lines 2-43 - compute user-specific text block scores for each text block in a document. The trained neural network may then be used to predict labels such as "of user interest" or "not of user interest" for each text block in a new document based on these viewing times (or other measurements of viewing behavior). Set of user-specific focus scores that ranges between 0.0 (indicating no interest) and 1.0 (indicating high interest) for each of a training set of text blocks may be used for training a neural network. determine an initial set of focus scores and determine a loss function value by taking a difference between the initial set of focus scores and the reference outputs. The system may propagate the loss function value through the neurons of the neural network and re-calculate the neural network parameters using gradient descent; col. 13, lines 23-37 - output of the neural network used may be a label, such as "of interest to user" or "not of interest to user," and the user specific criterion may be based on the label determined by the neural network. output of the neural network may be a quantitative value used to determine whether the text block score satisfies a user-specific display criterion. system may determine that a user-specific display criterion having an associated threshold value of 50% is satisfied if a user-specific text block score determined by the neural network is greater than 50%.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Guo, to include duration prediction model ... duration prediction model ... updated duration prediction model for the purpose of efficiently determining prediction scores based on view or interaction time period for determining loss values and used in model training, improving model accuracy, as taught by Walters (col. 12).
Claim(s) 13 and 19:
Claim(s) 13 and 19 correspond to Claim 6, and thus, Guo and Walters teach or suggest the limitations of claim(s) 13 and 19 as well.
Claim 7:
Guo further teaches or suggests inputting the target account and the target content into the content recommendation model to obtain the probability prediction result of the target content; determining a target recommendation content from the target content based on the probability prediction result of the target content; and pushing the target recommendation content to the target account (see Fig. 2, 5-8; para. 0196 - a probability that a to-be-processed user selects a candidate recommended object in a candidate recommended object set may be predicted by inputting a user behavior log of the to-be-processed user and the candidate recommended object set into a pre-trained recommendation model. The pre-trained recommendation model may be used to perform online inference on a probability that the user selects a recommended commodity based on interests and hobbies of the user; para. 0214 - candidate recommended object in the candidate recommended object set may be arranged based on a predicted probability that the user selects the candidate recommended object, to obtain a recommendation result of the candidate recommended object; para. 0216 – probabilities that a user selects the commodities in the candidate set, and arranges the candidate commodities in descending order of the probabilities, to arrange, at the most front, an application that is most likely to be downloaded. para. 0229 - input the user characteristic information, the context information, and the candidate recommended object set into a pre-trained recommendation model to obtain a probability that the to-be-processed user selects a candidate recommended object in the candidate recommended object set, where the pre-trained recommendation model is used to predict, when the user pays attention to a target recommended commodity, a probability that the user selects the target recommended object; and obtain a recommendation result of the candidate recommended object based on the probability that the to-be-processed user selects the candidate recommended object, where a model parameter of the pre-trained recommendation model is obtained by performing joint training; para. 0230 – candidate recommended object in the candidate recommended object set may be arranged based on a predicted probability that the user selects the candidate recommended object, to obtain a recommendation result of the candidate recommended object.).
Claim(s) 14 and 20:
Claim(s) 14 and 20 correspond to Claim 7, and thus, Guo and Walters teach or suggest the limitations of claim(s) 14 and 20 as well.
Claim(s) 2, 3, 9, 10, and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Guo, in view of Walters, and further in view of Sainz et al., US Publication 20220036542 (“Sainz”).
Claim 2:
Guo further teaches or suggests wherein the interaction data between the historical account and the historical recommendation content comprises a historical selection relationship between the historical account and the historical recommendation content, and historical view ... of the historical recommendation content by the historical account; and the determining the probability prediction loss corresponding to the probability prediction result ... comprise: determining the probability prediction loss based on the probability prediction result and the historical selection relationship; ... and determining ... of the probability prediction loss ... to obtain a prediction loss (see para. 0092 - updated based on a difference between the current predicted value and the target value. loss function (loss function) or an objective function (objective function). The loss function and the objective function are important equations used to measure the difference between the predicted value and the target value; para. 0136 - performing joint training on a ... model and a recommendation model by using the sample user behavior log and ... information of the sample recommended object as input data and using the sample label as a target output value, to obtain a trained recommendation model; para. 0165 - based on a difference between the actual sample label and a jointly predicted selection probability. the model parameters of the position aware model and the recommendation model may be obtained through a plurality of iterations based on the difference between the sample label and the jointly predicted selection probability; para. 0166 - predicted selection probability may be a predicted probability that includes ... that the user selects the sample object; para. 0211 - training may be training model parameters ... the recommendation model based on a difference between the sample label and a jointly predicted selection probability, and the jointly predicted selection probability is obtained based on output data of the ... model and the recommendation model.).
Walters further teaches or suggests ... view duration ...; ... and duration prediction loss corresponding to the duration prediction result ...; determining the duration prediction loss based on the duration prediction result and the historical view duration; ... and the duration prediction loss (see col. 11, lines 27-41 - Metrics of user focus may include the amount of time that the text block stays on the screen (e.g., the amount of time being greater than or equal to a threshold amount of time or otherwise satisfying the threshold amount. user-specific focus score may range from 0.0 to 1.0; col. 12, lines 2-43 - compute user-specific text block scores for each text block in a document. The trained neural network may then be used to predict labels such as "of user interest" or "not of user interest" for each text block in a new document based on these viewing times (or other measurements of viewing behavior). Set of user-specific focus scores that ranges between 0.0 (indicating no interest) and 1.0 (indicating high interest) for each of a training set of text blocks may be used for training a neural network. determine an initial set of focus scores and determine a loss function value by taking a difference between the initial set of focus scores and the reference outputs. The system may propagate the loss function value through the neurons of the neural network and re-calculate the neural network parameters using gradient descent; col. 13, lines 23-37 - output of the neural network used may be a label, such as "of interest to user" or "not of interest to user," and the user specific criterion may be based on the label determined by the neural network. output of the neural network may be a quantitative value used to determine whether the text block score satisfies a user-specific display criterion. system may determine that a user-specific display criterion having an associated threshold value of 50% is satisfied if a user-specific text block score determined by the neural network is greater than 50%.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Guo, to include ... view duration ...; ... and duration prediction loss corresponding to the duration prediction result ...; determining the duration prediction loss based on the duration prediction result and the historical view duration; ... and the duration prediction loss for the purpose of efficiently determining prediction scores based view or interaction time period for determining loss values and used in model training, improving model accuracy, as taught by Walters (col. 12).
Guo does not explicitly disclose a weighted sum.
Sainz teaches or suggests a weighted sum (see para. 0032 - combined weighted objective function may be a weighted sum of the classification loss function, attention loss function, and localization loss function. the classification loss function may be multiplied by a first scalar multiplier, and the localization loss function may be multiplied by a second scalar multiplier, and the attention loss function may be multiplied by a third scalar multiplier. The first, second, and third scalar multipliers may be values which result in the most accurate training of the AI system to perform its combined tasks of classification. first, second, and third scalar multiplier may be determined manually or by a computer by trial and error (e.g., testing possible values [using a random, or pseudo-random, number generator] to determine which values produce the most accurate results.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Guo, to include a weighted sum for the purpose of efficiently training a model by jointly training sub-models and generating a weighted sum of the losses from the sub-models to more effectively minimize overall loss, improving model accuracy, as taught by Sainz (para. 0032).
Claim(s) 9 and 16:
Claim(s) 9 and 16 correspond to Claim 2, and thus, Guo, Walters, and Sainz teach or suggest the limitations of claim(s) 9 and 16 as well.
Claim 3:
Guo further teaches or suggests determining ... the probability prediction loss and a probability (see para. 0092 - updated based on a difference between the current predicted value and the target value. loss function (loss function) or an objective function (objective function). The loss function and the objective function are important equations used to measure the difference between the predicted value and the target value; para. 0136 - performing joint training on a ... model and a recommendation model by using the sample user behavior log and ... information of the sample recommended object as input data and using the sample label as a target output value, to obtain a trained recommendation model; para. 0165 - based on a difference between the actual sample label and a jointly predicted selection probability. the model parameters of the position aware model and the recommendation model may be obtained through a plurality of iterations based on the difference between the sample label and the jointly predicted selection probability; para. 0166 - predicted selection probability may be a predicted probability that includes ... that the user selects the sample object; para. 0211 - training may be training model parameters ... the recommendation model based on a difference between the sample label and a jointly predicted selection probability, and the jointly predicted selection probability is obtained based on output data of the ... model and the recommendation model.).
Walters further teaches or suggests determining ... the duration prediction loss and a duration (see col. 11, lines 27-41 - Metrics of user focus may include the amount of time that the text block stays on the screen (e.g., the amount of time being greater than or equal to a threshold amount of time or otherwise satisfying the threshold amount. user-specific focus score may range from 0.0 to 1.0; col. 12, lines 2-43 - compute user-specific text block scores for each text block in a document. The trained neural network may then be used to predict labels such as "of user interest" or "not of user interest" for each text block in a new document based on these viewing times (or other measurements of viewing behavior). Set of user-specific focus scores that ranges between 0.0 (indicating no interest) and 1.0 (indicating high interest) for each of a training set of text blocks may be used for training a neural network. determine an initial set of focus scores and determine a loss function value by taking a difference between the initial set of focus scores and the reference outputs. The system may propagate the loss function value through the neurons of the neural network and re-calculate the neural network parameters using gradient descent; col. 13, lines 23-37 - output of the neural network used may be a label, such as "of interest to user" or "not of interest to user," and the user specific criterion may be based on the label determined by the neural network. output of the neural network may be a quantitative value used to determine whether the text block score satisfies a user-specific display criterion. system may determine that a user-specific display criterion having an associated threshold value of 50% is satisfied if a user-specific text block score determined by the neural network is greater than 50%.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Guo, to include determining ... the duration prediction loss and a duration for the purpose of efficiently determining prediction scores based view or interaction time period for determining loss values and used in model training, improving model accuracy, as taught by Walters (col. 12).
Saiz further teaches or suggests ... a product of ... weight parameter to obtain a first weight part; ... a product of ... weight parameter to obtain a second weight part; and determining a sum of the first weight part and the second weight part as the prediction loss, the probability weight parameter and the duration weight parameter being preset parameters (see para. 0032 - combined weighted objective function may be a weighted sum of the classification loss function, attention loss function, and localization loss function. the classification loss function may be multiplied by a first scalar multiplier, and the localization loss function may be multiplied by a second scalar multiplier, and the attention loss function may be multiplied by a third scalar multiplier. The first, second, and third scalar multipliers may be values which result in the most accurate training of the AI system to perform its combined tasks of classification. first, second, and third scalar multiplier may be determined manually or by a computer by trial and error (e.g., testing possible values [using a random, or pseudo-random, number generator] to determine which values produce the most accurate results.).
Accordingly, it would have been obvious to one having ordinary skill before the effective filing date of the claimed invention to modify the system and method, taught in Guo, to include ... a product of ... weight parameter to obtain a first weight part; ... a product of ... weight parameter to obtain a second weight part; and determining a sum of the first weight part and the second weight part as the prediction loss, the probability weight parameter and the duration weight parameter being preset parameters for the purpose of efficiently training a model by jointly training sub-models and generating a weighted sum of the losses from the sub-models to more effectively minimize overall loss, improving model accuracy, as taught by Sainz (para. 0032).
Claim(s) 10:
Claim(s) 10 correspond to Claim 3, and thus, Guo, Walters, and Sainz teach or suggest the limitations of claim(s) 10 as well.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Andrew T McIntosh whose telephone number is (571)270-7790. The examiner can normally be reached M-Th 8:00am-5:30pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Tamara Kyle can be reached at 571-272-4241. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ANDREW T MCINTOSH/Primary Examiner, Art Unit 2144