DETAILED ACTION
This Office Action is responsive to Request for Continued Examination filed on January 23rd, 2026. Claims 1 and 14 have been amended; claims 1-20 are pending and have been examined.
Any previous objections/rejections not mentioned in this action have been withdrawn by the Examiner.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendments and Arguments
Regarding rejections made under 35 U.S.C. 103, Applicant argues, "… the Office Action has failed to establish a prima facie case of obviousness," (page 7 of Remarks) and "… a claimed invention combining multiple known elements is not rendered obvious simply because each element was known independently in the prior art," (page 8 of Remarks).
Applicant further argues "that Freitas does not disclose 'identifying, by the ambiguity detection engine, a reason why the new data is ambiguous in meaning.' Instead, Freitas determines whether a grammatical reference - not the meaning of the data - is ambiguous… Thus, Freitas does not consider the meaning of the data, only whether a grammatical reference is unclear," (emphasis original, page 9 of Remarks) and further, "Freitas does not replace the data, but instead modifies the reference to the data… Thus, instead of replacing new data as claimed, Freitas discloses replacing the reference involving the data," (page 10 of Remarks).
Applicant’s argument has been considered, but is not persuasive. Regarding Freitas’ teachings of ambiguity, the argument shows a distinction without difference; Freitas’ Background teaches that “[w]ords may be unambiguous at their face value, but may have different meanings, subtleties, or sub-groups that affect their meaning…
For example, if it is known that a document's content is referring to the topic of Education, it is known that the word ‘school’ is more likely referring to the sense of an ‘educational institution’ rather than the meaning of a ‘group of fish’. This is where traditional word-sense disambiguation processes would stop, satisfied that the job is finished. However, a lot of ambiguity still remains. Is it a secondary or primary school, or is it even a school in Ireland? The word ‘school’ is inherently ambiguous because it does not provide enough information. Perhaps the school in question is actually for training dogs rather than people.
The term ‘latent natural language ambiguity’ is used herein to describe this phenomenon, borrowing from the legal definition of ‘latent ambiguity’ where the wording of an instrument is on the face of it clear and intelligible, but may, at the same time, apply equally to two different things or subject matters.”
By resolving matters of unclear grammatical reference, Freitas is teaching a method that handles data with unclear or ambiguous meaning. The claims, under broadest reasonable interpretation, offer no limitations to distinguish a method to detect “that the new data is ambiguous” and Freitas’ method to detect latent ambiguity.
Regarding the replacement of ambiguous data, Freitas “The method may include replacing any annotated nodes indicating a latent ambiguity of the concept remaining in an output of the processing of the natural language input with an alternative node from the semantic graph,” (paragraph [0008]. The claims offer no limitation to distinguish “a modification to the document by replacing the new data with unambiguous data that addresses the reason” from Freitas’ methods of replacing ambiguous nodes in a semantic graph with more appropriate alternatives.
Regarding a motivation to combine Quinton and Freitas, the MPEP teaches, in section 2144, that the “strongest rationale for combining references is a recognition, expressly or impliedly in the prior art or drawn from a convincing line of reasoning based on established scientific principles or legal precedent, that some advantage or expected beneficial result would have been produced by their combination.” Freitas is used in combination with Quinton to teach the limitations of the claims pertaining to a reasoned identification of ambiguous data and the subsequent modification to correct the data, thereby improving the both the readability of the original document as well as the user experience as a benefit of automation.
Freitas teaches that “[u]sers of computing devices often use various content creation applications to create textual content. For example, users may utilize an application to write an email, prepare an essay, document their work, prepare a presentation and the like. Proper use of references in content is important for creating unambiguous and accurate content,” which aligns with the motivation of the Specification, “Large organizations often have offices and staff across many continents who serve vital interconnected business functions. This necessitates vast quantities of globe-spanning communication between people who may not share a first language. Unfortunately, internal documentation can be confusing, unclear or inadequate.”
Quinton and Freitas are each concerned with the identification of ambiguous data in the context of language processing; therefore, a person having skill in the art and a motivation to improve a user-facing system for performing the method taught by Quinton, would have reasonably included the features taught by Freitas to improve the user experience by both identifying ambiguity as well as replacing offending elements (see MPEP 2144.04.III).
In accordance with the above reasoning, the rejections under 35 U.S.C. 103 are maintained. Further details are provided below.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-6, 8, 10, 14, 16 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Application Publication 2022/0019944 to Quinton et al. (hereinafter, "Quinton") in view of U.S. Patent 11,397,846 to Freitas (hereinafter, “Freitas”).
Regarding claim 1, Quinton teaches a method for neurodivergence-driven ambiguity detection and resolution, comprising: receiving, by an ambiguity detection training computer program executed by an electronic device, a labeled dataset comprising data labeled by one or more neurodivergent individuals, wherein the data is labeled as clear or ambiguous in meaning (paragraph [0119], "Embodiments of ambiguous label output annotation may include utilizing a human expert to analyze the training data and identify elements of the training data for annotating. For example, the analysis may include identifying ambiguous input data and annotating accordingly.");
training, by the ambiguity detection training computer program, an ambiguity detection engine using the labeled dataset to predict ambiguities in new data (paragraph [0076], "Machine learning systems are capable of learning to predict labels based on training processes having training data comprising inputs and associated labels. Subsequent testing processes may be used to validate the efficacy of the training process before deploying the machine learning system. In a deployed state, the machine learning system predicts a label for a given input on the basis of the configuration conditioned by the training process.");
deploying, by the ambiguity detection training computer program, the ambiguity detection engine to a computer program or system (paragraph [0084], "In the deployment scenario, the trained machine learning engine 130 receives an input 182 from an external source 180, such as a higher level system, sensor, or data file. The machine learning engine 130 applies the parameters 140 to the input 182 to generate an output 134 which inputs to the prediction engine 160 for generating a predicted label 164. In the deployed scenario, there is no way to known whether the predicted label is correct or not.");
receiving, by the ambiguity detection engine a document comprising the new data (paragraph [0084], " In the deployment scenario, the trained machine learning engine 130 receives an input 182 from an external source 180, such as a higher level system, sensor, or data file."); and
predicting, by the ambiguity detection engine, that the new data is ambiguous in meaning (paragraph [0084], "The machine learning engine 130 applies the parameters 140 to the input 182 to generate an output 134 which inputs to the prediction engine 160 for generating a predicted label 164.").
Quinton does not explicitly teach “identifying, by the ambiguity detection engine, a reason why the new data is ambiguous in meaning,” “generating, by the ambiguity detection engine, a modification to the document by replacing the new data with unambiguous data that addresses the reason,” or “outputting, by the ambiguity detection engine, the document comprising the modification,” and thus, Freitas is introduced.
Freitas teaches identifying, by the ambiguity detection engine, a reason why the new data is ambiguous in meaning (column 12, lines 34-38, "The UI element 240 may provide an explanation such as the description 245 that identifies an identified ambiguous reference, explains why the reference is ambiguous and/or provides a reason for the suggested reference being more appropriate.");generating, by the ambiguity detection engine, a modification to the document by replacing the new data with unambiguous data that addresses the reason (column 9, lines 43-51, "Once the improper references are identified, they may be transmitted to one or more reference suggestion models 152. The reference suggestion models 152 may examine the identified improper reference(s) and provide suggestions for replacing the improper reference(s) with appropriate ones. This may include examining an improper reference, determining if the reference is ambiguous, incorrect or inconsistent, and then identifying one or more proper references that can be used to replace the improper reference."); andoutputting, by the ambiguity detection engine, the document comprising the a modification (column 12, lines 55-58, "The user may be able to automatically replace the identified improper reference by selecting one of the suggested references (e.g., by clicking on one of the suggested references).").
Quinton and Freitas are considered analogous because they are each concerned with machine learning against ambiguous data. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified Quinton with the teachings of Freitas for the purpose of improving model usability. Given that all the claimed elements were known in the prior art, one skilled in the art could have combined the elements by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results.
Regarding claim 2, Quinton further teaches The method of claim 1, wherein the data comprises text or graphics (paragraph [0056], "According to an aspect, the present disclosure provides an ambiguity-aware machine learning system for identifying an object, comprising a user device having a sensor for acquiring information indicative of the object, the user device communicatively coupled to a processor configured by machine-readable instructions to generate image data based on the information acquired by the sensor, the image data indicative of a first perspective of the object; ").
Regarding claim 3, Quinton further teaches The method of claim 1, wherein the data is labeled on a scale (paragraph [0079], “Conventionally, training data comprises a plurality of training elements, each training element having an input data and an associated label or desired label. Examples of labels include numeric or symbolic values. For example, the label may be a “one-hot” encoded vector with a length equal to the number of valid labels, with each position in the vector being used to represent each different label such that a value of ‘1’ in the position corresponding to a specific label and values of ‘0’ in all other locations represents another specific label. Many other label definitions are possible.”).
Regarding claim 4, Quinton further teaches The method of claim 1, wherein the data is labeled with a replacement (paragraph [0070], "In an embodiment, further annotating the set of training data to include a plurality of desired responses correspondingly associated with the plurality of input data.").
Regarding claim 5, Quinton further teaches The method of claim 1, wherein the ambiguity detection engine is trained using supervised learning (paragraph [0118], "Embodiments of an ambiguity-aware machine learning system as disclosed herein may include applying a supervised learning training process to the ambiguous label output by including a desired response for the output in the set of training data.").
Regarding claim 6, Quinton teaches The method of claim 1, further comprising: verifying, by the ambiguity detection training computer program, the ambiguity detection engine before it is deployed (paragraph [0076], "Subsequent testing processes may be used to validate the efficacy of the training process before deploying the machine learning system. In a deployed state, the machine learning system predicts a label for a given input on the basis of the configuration conditioned by the training process.").
Regarding claim 8, Quinton further teaches The method of claim 1, wherein the ambiguity detection engine is available via an Application Programming Interface (paragraph [0163], "The mass storage device 1110 may comprise any type of storage device configured to store data, programs, and other information and to make the data, programs, and other information accessible via the bus 1112. The mass storage device 1110 may comprise one or more of a solid state drive, hard disk drive, a magnetic disk drive, an optical disk drive, or the like. In some embodiments, data, programs, or other information may be stored remotely, for example in the cloud. Computerized system 1100 may send or receive information to the remote storage in any suitable way, including via communications subsystem 1108 over a network or other data communication medium.").
Regarding claims 10, Quinton teaches The method of claim 1, further comprising: receiving, by the ambiguity detection engine, feedback on the modification (paragraph [0121], "Embodiments of an ambiguous label output annotation process may apply iteratively, gradually annotating desired responses of the ambiguous label output."); andre-training, by the ambiguity detection training computer program, the ambiguity detection engine based on the feedback (paragraph [0121], "Embodiments of an ambiguous label output annotation process may apply iteratively, gradually annotating desired responses of the ambiguous label output. In an embodiment, the annotating process includes first annotating all training data as non-ambiguous, followed by performing the training process and analyzing failure patterns in the training data, and/or test data, and/or validation data. Based on these failure patterns, the desired response of the ambiguous label output for some subset of the data can be updated to indicate that they should be considered ambiguous. This process is repeated starting at the application of the training process.”).
Regarding claim 14, Quinton teaches a system, comprising: an electronic device executing an ambiguity detection training computer program (paragraph [0159], “FIG. 11 is a block diagram of an example computerized device or system 1100 that may be used in implementing one or more aspects or components of an embodiment of an ambiguity-aware machine learning system according to the present disclosure.”);
a labeled dataset comprising data labeled by one or more neurodivergent individuals, wherein the data is labeled as clear or ambiguous (paragraph [0034], “In an embodiment, performing a training process by the machine learning engine utilizing training data and a cost function to determine the set of internal parameters, wherein the training data utilized in the training process includes a subset of the training data that includes ambiguity indications indicating that the training inputs in the subset of training data may be considered ambiguous,” and paragraph [0119], "Embodiments of ambiguous label output annotation may include utilizing a human expert to analyze the training data and identify elements of the training data for annotating. For example, the analysis may include identifying ambiguous input data and annotating accordingly."); and
a user electronic device executing a computer program (paragraph [0056], "According to an aspect, the present disclosure provides an ambiguity-aware machine learning system for identifying an object, comprising a user device having a sensor for acquiring information indicative of the object…");
wherein: the ambiguity detection training computer program receives the labeled dataset (paragraph [0082], “FIG. 1 illustrates a training process for a previous machine learning system 100 which trains on a set of training data 110. The training data 110 consists of a plurality of training elements, namely a plurality of input data 112a, . . . , 112y, 112z paired with a corresponding plurality of labels 114a, . . . , 114y, 114z. The machine learning system 100 includes a machine learning engine 130 having an associated set of activations 132 and associated set of parameters 140. During the training process, the machine learning engine 130 receives an input 112 corresponding to one of the plurality of input data 112a, . . . , 112y, 112z.”);
the ambiguity detection training computer program trains an ambiguity detection engine using the labeled dataset to predict ambiguities in meaning of new data (paragraph [0113], "For example, the supervised learning aspect relates to using the known correct labels to train machine learning system outputs in the case that the input data is not deemed ambiguous; while the unsupervised learning aspect relates to training which input data is ambiguous under the notion that only a subset of the training data should be allowed to be considered as ambiguous.");
the ambiguity detection training computer program deploys the ambiguity detection engine to the computer program (paragraph [0084], “FIG. 3 illustrates a deployment scenario for a previous machine learning system 100 having a machine learning engine 130 trained in accordance with FIG. 1 and optionally tested in accordance with the process of FIG. 2. In the deployment scenario, the trained machine learning engine 130 receives an input 182 from an external source 180, such as a higher level system, sensor, or data file.”); and
the computer program receives a document comprising the new data (paragraph [0084], “In the deployment scenario, the trained machine learning engine 130 receives an input 182 from an external source 180, such as a higher level system, sensor, or data file.”);
the computer program predicts, using the ambiguity detection engine, that the new data is ambiguous in meaning (paragraph [0088], "For a given input, the machine learning system disclosed herein provides an ambiguity-aware prediction which associates a suitable label (from a set of supported labels) or an ambiguous label with the given input data. The ambiguous label provides an indication of ambiguity in the given input data. The ambiguous label may be provided as a separate output or encoded in a prediction output with the suitable label from the set of supported labels.").
Quinton does not explicitly teach “the computer program identifies, using the ambiguity detection engine, a reason why the new data is ambiguous in meaning,” “the computer program generates, using the ambiguity detection engine, a modification to the document by replacing the new data with unambiguous data that addresses the reason,” or “the computer program outputs the document comprising the modification,” and thus, Freitas is referenced.
Freitas teaches the computer program identifies, using the ambiguity detection engine, a reason why the new data is ambiguous in meaning (column 12, lines 34-38, "The UI element 240 may provide an explanation such as the description 245 that identifies an identified ambiguous reference, explains why the reference is ambiguous and/or provides a reason for the suggested reference being more appropriate.");
the computer program generates, using the ambiguity detection engine, a modification to the document by replacing the new data with unambiguous data that addresses the reason (column 9, lines 43-51, "Once the improper references are identified, they may be transmitted to one or more reference suggestion models 152. The reference suggestion models 152 may examine the identified improper reference(s) and provide suggestions for replacing the improper reference(s) with appropriate ones. This may include examining an improper reference, determining if the reference is ambiguous, incorrect or inconsistent, and then identifying one or more proper references that can be used to replace the improper reference."); and
the computer program outputs the document comprising the modification (column 12, lines 55-58, "The user may be able to automatically replace the identified improper reference by selecting one of the suggested references (e.g., by clicking on one of the suggested references).").
Quinton and Freitas are considered analogous because they are each concerned with machine learning against ambiguous data. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified Quinton with the teachings of Freitas for the purpose of improving model usability. Given that all the claimed elements were known in the prior art, one skilled in the art could have combined the elements by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results.
Regarding claim 16, Quinton further teaches The system of claim 14, wherein the ambiguity detection engine is available via an Application Programming Interface (paragraph [0163], "The mass storage device 1110 may comprise any type of storage device configured to store data, programs, and other information and to make the data, programs, and other information accessible via the bus 1112. The mass storage device 1110 may comprise one or more of a solid state drive, hard disk drive, a magnetic disk drive, an optical disk drive, or the like. In some embodiments, data, programs, or other information may be stored remotely, for example in the cloud. Computerized system 1100 may send or receive information to the remote storage in any suitable way, including via communications subsystem 1108 over a network or other data communication medium.").
Regarding claim 18, Quinton teaches The system of claim 14, further comprising: receiving, by the ambiguity detection engine, feedback on the modification (paragraph [0121], "Embodiments of an ambiguous label output annotation process may apply iteratively, gradually annotating desired responses of the ambiguous label output."); andre-training, by the ambiguity detection training computer program, the ambiguity detection engine based on the feedback (paragraph [0121], "Embodiments of an ambiguous label output annotation process may apply iteratively, gradually annotating desired responses of the ambiguous label output. In an embodiment, the annotating process includes first annotating all training data as non-ambiguous, followed by performing the training process and analyzing failure patterns in the training data, and/or test data, and/or validation data. Based on these failure patterns, the desired response of the ambiguous label output for some subset of the data can be updated to indicate that they should be considered ambiguous. This process is repeated starting at the application of the training process.”).
Claims 7, 9, 15 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Quinton and Freitas as applied to claims 1 and 14 above, further in view of U.S. Patent Application Publication 2023/0153546 to Peleg et al. (hereinafter, "Peleg").
Regarding claims 7 and 15, the combination of Quinton and Freitas does not explicitly teach a method or system “wherein the ambiguity detection engine is a plug-in to a computer program, and the computer program comprises a word processing program, a coding program, an email program, and/or a messaging program,” and thus, Peleg is introduced.
Peleg teaches the ambiguity detection engine is a plug-in to a computer program, and the computer program comprises a word processing program, a coding program, an email program, and/or a messaging program (paragraph [062], "In some cases, the writing assistant may be incorporated with applications including, but not limited to, email editors, word processing programs, presentation applications, spreadsheet applications, PDF editors, etc.").
Quinton, Freitas and Peleg are considered analogous because they are each concerned with machine learning against ambiguous data. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified Quinton and Freitas with the teachings of Peleg for the purpose of expanding model utility. Given that all the claimed elements were known in the prior art, one skilled in the art could have combined the elements by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results.
Regarding claims 9 and 17, the combination of Quinton and Freitas does not explicitly teach a method or system “wherein the ambiguity detection engine is configured to scan a code repository and/or a document repository,” however, Peleg teaches the ambiguity detection engine is configured to scan a code repository and/or a document repository (paragraph [0374], “Various techniques may be used for identifying source text segments for use by the writing assistant tool in generating text output re-purposing the identified text segments within a new document. For example, documents including source text segments may be selected from an interface window that lists files in a directory. Files shown in a directory may be dragged and dropped into a project window in order to identify to the writing assistant documents for re-purposing.”).
Quinton, Freitas and Peleg are considered analogous because they are each concerned with machine learning against ambiguous data. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified Quinton and Freitas with the teachings of Peleg for the purpose of expanding model utility. Given that all the claimed elements were known in the prior art, one skilled in the art could have combined the elements by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results.
Claims 11-13 and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Quinton and Freitas as applied to claims 1 and 14 above, further in view of "Enforcing fairness using ensemble of diverse Pareto-optimal models" by Guardieiro et al. (hereinafter, "Guardieiro").
Regarding claim 11, the combination of Quinton and Freitas does not explicitly teach “The method of claim 1, wherein the training results in a plurality of mutually non-dominating models,” and thus, Guardieiro is introduced.
Guardieiro teaches the training results in a plurality of mutually non-dominating models (section 2 Related work, "A key characteristic of the ensemble techniques is the requirement for a pool of diverse models. The Pareto approach for multi-objective optimization is capable of generating such a pool. Moreover, the models are trained to have the optimal trade-offs between the objective functions in the training set."). The recited “non-dominating models” are read to include models in a Pareto distribution, and for that reason, the teachings of Guardieiro read on the claim.
Quinton, Freitas and Guardieiro are considered analogous because they are each concerned with machine learning and data processing. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified Quinton and Freitas with the teachings of Guardieiro for the purpose of optimizing model accuracy. Given that all the claimed elements were known in the prior art, one skilled in the art could have combined the elements by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results.
Regarding claim 12, Guardieiro further teaches The method of claim 11, wherein the ambiguity detection engine is configured to select a detection level and to select one of the mutually non-dominating models based on the selection (section 5.1 Model aggregation, "Select models above minimum performance in this step, we select the best R1 ≤ R models w.r.t. their validation accuracy—this step is necessary to ensure the consistency of the results obtained. If we did not impose such a restriction, it could result in models that are fair for the training and validation data but with too low accuracy to be a useful predictor.").
Quinton, Freitas and Guardieiro are considered analogous because they are each concerned with machine learning and data processing. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified Quinton and Freitas with the teachings of Guardieiro for the purpose of optimizing model accuracy. Given that all the claimed elements were known in the prior art, one skilled in the art could have combined the elements by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results.
Regarding claim 13, Guardieiro further teaches The method of claim 11, wherein each of the mutually non-dominating models has a different inclination to produce both true positive results and false positive results (section 6.1.4 Evaluation metrics. "We select those metrics to evaluate different aspects of a fair model. Accuracy measures the overall capacity to make correct classifications. Equal opportunity and demographic parity measure how far the classifier can make a similar classification for any group on two aspects: correctly classifying the beneficial outcome and achieving the beneficial outcome. Finally, the coefficient of variation measures how much the classifier deviates from delivering the beneficial outcome for all individuals.").
Quinton, Freitas and Guardieiro are considered analogous because they are each concerned with machine learning and data processing. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified Quinton and Freitas with the teachings of Guardieiro for the purpose of optimizing model accuracy. Given that all the claimed elements were known in the prior art, one skilled in the art could have combined the elements by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results.
Regarding claim 19, the combination of Quinton and Freitas does not explicitly teach “The system of claim 14, wherein the training results in a plurality of mutually non-dominating models, wherein each of the mutually non-dominating models has a different inclination to produce both true positive results and false positive results,” however, Guardieiro teaches the training results in a plurality of mutually non-dominating models, wherein each of the mutually non-dominating models has a different inclination to produce both true positive results and false positive results (section 2 Related work, "A key characteristic of the ensemble techniques is the requirement for a pool of diverse models. The Pareto approach for multi-objective optimization is capable of generating such a pool. Moreover, the models are trained to have the optimal trade-offs between the objective functions in the training set," and section 6.1.4 Evaluation metrics. "We select those metrics to evaluate different aspects of a fair model. Accuracy measures the overall capacity to make correct classifications. Equal opportunity and demographic parity measure how far the classifier can make a similar classification for any group on two aspects: correctly classifying the beneficial outcome and achieving the beneficial outcome. Finally, the coefficient of variation measures how much the classifier deviates from delivering the beneficial outcome for all individuals.").
Quinton, Freitas and Guardieiro are considered analogous because they are each concerned with machine learning and data processing. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to have modified Quinton and Freitas with the teachings of Guardieiro for the purpose of optimizing model accuracy. Given that all the claimed elements were known in the prior art, one skilled in the art could have combined the elements by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results.
Regarding claim 20, Guardieiro further teaches The system of claim 19, wherein the ambiguity detection engine is configured to select a detection level and to select one of the mutually non-dominating models based on the selection (section 5.1 Model aggregation, "Select models above minimum performance in this step, we select the best R1 ≤ R models w.r.t. their validation accuracy—this step is necessary to ensure the consistency of the results obtained. If we did not impose such a restriction, it could result in models that are fair for the training and validation data but with too low accuracy to be a useful predictor.").
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
U.S. Patent Application Publication 2019/0370696 to Ezen Can et al.
U.S. Patent Application Publication 2020/0065385 to Dreher et al.
U.S. Patent Application Publication 2020/0380991 to Ge et al.
U.S. Patent Application Publication 2021/0027771 to Hall et al.
U.S. Patent Application Publication 2021/0216707 to Srinivasan et al.
U.S. Patent Application Publication 2022/0276862 to Balasubramanian et al.
U.S. Patent Application Publication 2022/0374784 to Sharma et al.
U.S. Patent Application Publication 2023/0072171 to Sengupta et al.
U.S. Patent 11,281,855 to Archuleta.
U.S. Patent 11,501,111 to Qian et al.
"Multi-objective optimisation in the presence of uncertainty” by Fieldsend and Everson.
"Multi-objective learning of Relevance Vector Machine classifiers with multi-resolution kernels” by Clark and Everson
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SEAN T SMITH whose telephone number is (571)272-6643. The examiner can normally be reached Monday - Friday 8:00am - 5:00pm.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, PIERRE-LOUIS DESIR can be reached at (571) 272-7799. 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.
/SEAN THOMAS SMITH/Examiner, Art Unit 2659
/PIERRE LOUIS DESIR/Supervisory Patent Examiner, Art Unit 2659
Prosecution Insights