DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application is being examined under the pre-AIA first to invent provisions.
Response to Arguments
103 Rejection Arguments
Applicant asserts:
Applicant argues, on pages 10-11, “none of Anamandra, Olshansky, or Knoppert teaches or suggests generating a tuning matrix from a circumstance-based model, based on circumstantial factors sensed in real time, that dynamically modifies the relative weighted contributions of a set of separate empathy models in real time.”
Examiner response:
Applicant' s arguments with respect to claim(s) 1, 5, 11, 15, and 20 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Applicant asserts:
Applicant argues, on pages 11, “none of the references teaches or suggests hosting the empathy and circumstance models within a trusted execution environment such that their underlying values and interconnections cannot be directly accessed or queried and can only be interacted with to generate a tuning matrix.”
Examiner response:
Applicant' s arguments with respect to claim(s) 1, 5, 11, 15, and 20 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Applicant asserts:
Applicant argues, on pages 11, “none of the references teaches or suggests automatically controlling presentation characteristics of a notification rendered on a user interface, including the timing of presentation, the content, and display characteristics of a graphical user interface element such as size, color, position, or notification type, based on the highest-scoring biased output.”
Examiner response:
Examiner respectfully disagrees. Examiner interprets the limitation as the interface is automatically updated/controlled to render a notification on the interface, where the presentation characteristics include a timing of presentation of the notification, content of the notification, and one or more display characteristics of a graphical user interface element of the notification including at least one of a size, color, position, or notification type. Examiner maps these features to Anamandra [0083] [0086]. The cited paragraph shows that once activation of the application has occurred, the functionalities is capable of automatically generating a user interface to show content of the notification.
Applicant asserts:
Applicant argues, on pages 11, “Applicant submits that the proposed combination is not obvious. The references are directed to disparate problems, predicting employee burnout (Anamandra), scoring job candidates (Olshansky), and adjusting keyboard haptics from typing dynamics (Knoppert), and the Office Action does not articulate a sufficient rationale, grounded in the references themselves, for combining them to arrive at the specific architecture now claimed.”
Examiner response:
Applicant’s arguments with respect to claim(s) 1, 5, 11, 15, and 20 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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.
Claim(s) 1, 5, 11, 15, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Anamandra et al. (US 20230131099, Herein “Anamandra”) in view of Olshansky (US 20200311682) in view of Knoppert et al. (US 20210240283, Herein “Knoppert”) in view of Linton et al. (US 20210125051 A1, Herein “Linton”) in view of Kumar et al. (US 20220377844 A1, Herein “Kumar”)
Regarding claim 1, Anamandra teaches A system for controlling generation of one or more computer-generated insights using empathy-based machine learning features (psychological modeling of user (abstract) performed on system (figs. 1 and 2)), the system comprising:
a processor, operating in conjunction with computer memory and data storage, the processor configured (processor, system [0002]) to:
maintain, in a first set of machine learning models each tracking an empathy-based aspect of a user (e.g., 135a and 135b (fig. 1)), a trained empathy based representation of the user , each of the machine learning models of the first set of machine learning models tracking a different empathy-based aspect of the user (respective user models, e.g., 135a and 135b (fig. 1) each model trained on, e.g., indicators or notifications of employee burnout [0045]; Examiner’s note: Olshansky more specifically teaches empathy), [wherein the first set of machine learning models is hosted within a trusted execution environment such that underlying values and interconnections of the first set of machine learning models cannot be directly accessed or queried and can only be interacted with to generate a tuning matrix;]
maintain, in a second machine learning model [hosted within the trusted execution environment], a trained circumstance based representation of the user, receive one or more candidate insight data objects representing potential computer- based interactive notifications (a global ML model 115 (fig. 1); receiving parameter spaces of user such as receiving at least a first parameter space [0006]; model 115 trained using personal data [0046] representing user behaviors based on, e.g., detection and leading indicators [0051]);
generate a tuning matrix from the second machine learning model, [based on circumstantial factors sensed in real time], to be applied as biasing weights to the trained empathy based representation of the user [to dynamically modify relative weighted contributions of each of the different machine learning models of the first set of machine learning models in real time]; (updated global machine learning model parameter space [0007]; update the trained empathy based representation of the user (e.g., local machine learning model 135a) [0060]; the updating the parameter space based on weight propagation [0049]; the at, the parameter space passage includes weights for model update [0060]);
process each of the one or more candidate insight data objects using at least the biasing weights applied to the trained empathy based representation of the user to generate a real-time prediction score for each of the one or more candidate insight data objects (at the local machine learning model, based on the updated parameter/weight space [0039] process user activity based on a respective selected machine learning model to calculate a prediction corresponding to a calculated output value of a respective user model, such as involving a confidence score for, e.g., various points in time and further variables [0048]; in particular, root causes corresponding with a determined top k quantity of root causes [0048]; further, the local machine learning model 135 performs inference [0049]); and
transmit the candidate insight data object having a highest score to a user interface associated with the user (determined top k quantity of root causes which may are aggregated and shared with an employer (i.e., the employer associated with the user) [0048]).
And automatically control, based on the candidate insight data object having the highest score, one or more presentation characteristics of a notification rendered on the user interface, the one or more presentation characteristics including a timing of presentation of the notification, content of the notification, and one or more display characteristics of a graphical user interface element of the notification including at least one of a size, color, position, or notification type. (The processor 710 is capable of processing instructions stored in the memory 720 and/or on the storage device 730 to display graphical information (ie. The top k quantity of root causes) for a user interface provided via the input/output device 740. [0083]; Upon activation within the applications, the functionalities can be used to generate the user interface provided via the input/output device 740. The user interface can be generated and presented to a user by the computing system 700 (e.g., on a computer screen monitor, etc.). [0086])
However, while Anamandra discloses trained machine learning models to predict user behavior based on application to a user corresponding to such psychological data as employee burnout (abstract), Anamandra fails to specifically teach machine learning models each tracking an empathy-based aspect of a user.
Yet, in a related art, Olshansky discloses candidate empathy models based on a user role [0119] such that each empathy model is based on analyzed behavioral data [0004].
It would have been obvious to one of ordinary skill in the art prior to the invention’s effective filing date to combine the empathy model of Olshansky with the personality model of Anamandra to have machine learning models each tracking an empathy-based aspect of a user. The combination would allow for, according to the motivation of Olshansky, automating content interaction [0001] by examining behavioral data of the user particularly with respect to empathy for better determining representative content for the user [0002]; e.g., the empathy score model is generated by observing user behavior such that the model can be used to determining a user empathy and apply to candidates in a candidate database [0004] to more effectively determine content for a user [0006].
However, while Anamandra discloses update the trained empathy based representation of the user (e.g., local machine learning model 135a) [0060], Anamandra in view of Olshansky fails to specifically teach tuning matrix.
Yet, in a related art, Knoppert discloses machine learning model modeled using weight matrices [0117] and, in particular, a training performed based on adjustments according to the weight matrices [0118] for fine tuning the weight matrices [0119].
It would have been obvious to one of ordinary skill in the art prior to the invention’s effective filing date to combine the tuning matrix of Knoppert with the empathy modeling and processing of Anamandra in view of Olshansky to have tuning matrix. The combination would allow for, according to the motivation of Knoppert, adjusting according to weight matrix in order to more accurately reflect model parameters that describe the individual user in a particular mood state [0118] and [0119].
However, while Anamandra discloses maintaining a first set 135a and 135b (fig. 1) and second machine learning model(s) a global ML model 115 (fig. 1), Anamandra fails to specifically teach machine learning models that are hosted within a trusted execution environment.
Yet, in a related art, Linton discloses hosting machine learning models within a trusted execution environment [0034] and, in particular, the machine learning models can be only be interacted via API [0039].
It would have been obvious to one of ordinary skill in the art prior to the invention’s effective filing date to combine the trusted execution environment of Linton with the empathy modeling and processing of Anamandra in view of Olshansky in further view of Knoppert to have the models hosted within the trusted execution environment. The combination would allow for, according to the motivation of Knoppert, fine control of the DNN model without worry of execution from malicious actors [0028]
However, while Anamandra discloses updated global machine learning model parameter space [0007]; the updating the parameter space based on weight propagation [0049]; the at, the parameter space passage includes weights for model update [0060], Anamandra fails to specifically teach updating the tuning matrix based on circumstantial factors sensed from the different machine learning models in real time to modify the relative weighted contributions
Yet, in a related art, Kumar discloses updating centralized model repository based UE/NE generate/collected real-time data [0083] from the different UEs where weight/updates are aggregated
It would have been obvious to one of ordinary skill in the art prior to the invention’s effective filing date to combine the ML model training procedure of Kumar with the empathy modeling and processing of Anamandra in view of Olshansky in further view of Knoppert in further view of Linton to train the empathy-based models using data from different machine learning models. The combination would allow for, according to the motivation of Kumar, enable the model to learn an optimal policy that maximizes reward [0074] and adjust weights to reduce errors of the output [0080]
Regarding claim 5, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar teaches the limitations of claim 1, as above.
Furthermore, Anamandra teaches The system of claim 1, wherein the second machine learning model is configured to track environmental features associated with a particular contextual environment of the user (for updating the global machine learning model [0039], associated with contextual environment of the user such as causes for an employee decline [0052]).
Regarding claim 11, the claim recites similar limitations as claim 1 – see above.
Regarding claim 15, the claim recites similar limitations as claim 5 – see above.
Regarding claim 20, Anamandra teaches A non-transitory computer readable medium storing machine interpretable instruction sets, which when executed by a processor, cause the processor to perform a method for controlling generation of one or more computer-generated insights using empathy-based machine learning features (machine learning model (abstract) on a system [0002], figs. 1 and 2), the method comprising:
The claim recites similar limitations as claim 1 – see above.
Claim(s) 2, 9, 12, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Anamandra in view of Olshansky in view of Knoppert in view of Linton in view of Kumar in view of Nagarajan et al. (US 11,593,677, Herein “Nagarajan”).
Regarding claim 2, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar teaches the limitations of claim 1, as above.
However, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar fails to specifically teach The system of claim 1, wherein the processor is further configured to: receive, from the user interface associated with the user, a data set representative of an outcome associated with presentation of the candidate insight data object to the user; and
re-train the first set of machine learning models and the second machine learning model using the data set representative of the outcome associated with presentation of the candidate insight data object to the user.
Yet, in a related art, Nagarajan discloses based on a user selection performing retraining of the first and second machine learning models (fig. 6).
It would have been obvious to one of ordinary skill in the art prior to the invention’s effective filing date to combine the retraining of the first and second models and receiving data associated with the presentation of candidate object from the user of Nagarajan with the empathy based modeling and execution of Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar to have receive, from the user interface associated with the user, a data set representative of an outcome associated with presentation of the candidate insight data object to the user; and re-train the first set of machine learning models and the second machine learning model using the data set representative of the outcome associated with presentation of the candidate insight data object to the user. The combination would allow for, according to the motivation of Nagarajan, retraining the machine learning models based on a user selection particularly involving a user interaction with a user interface of a device for optimizing assets or objects that a user can interact with (fig. 6 and corresponding specification discussion) thus improving the generation of content for the user based not only on a better prediction of an aspect of a user using categorizqation machine learning model and a user activity profile, but also learning based on the user activity associated with the user (i.e., retraining) (cols. 1 and 2).
Regarding claim 9, Anamadra in view of Olshansky in view of Knoppert in view of Nagarajan teaches the limitations of claims 1 and 2, as above.
Furthermore, Nagarajan teaches The system of claim 2, wherein the data set representative of the outcome includes interactions on the user interface associated with the notification (model training corresponding with user historical data such as activity profile for prediction (figs. 3 and 4)).
Regarding claim 12, the claim recites similar limitations as claim 2 – see above.
Regarding claim 19, the claim recites similar limitations as claim 9 – see above.
Claim(s) 3 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Anamandra in view of Olshansky in view of Knoppert in view of Linton in view of Kumar in view of Chandrasekara (US 20190266999, Herein “Chandrasekara”).
Regarding claim 3, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar teaches the limitations of claim 1, as above.
However, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar fails to specifically teach The system of claim 1, wherein the empathy-based aspects include at least one of curiosity, preconceptions, inspirations, direct experience, listening, or imagination.
Yet, in a related art Chandrasekara discloses user direct experiences [0004] and further user’s emotions such as somber, happy, good/bad, past interactions, frustrated, relaxed, happy, hurried, polite, matter of fact, heart rate, perspirations, et. [0034] to [0043] corresponding with data to be used with machine learning algorithms to learn mapping between a situation and most appropriate emotional reaction for the user [0005].
It would have been obvious to one of ordinary skill in the art prior to the invention’s effective filing date to combine any of the curiosity, preconceptions, inspirations, direct experience, listening, or imagination emotional characteristics of Chandrasekara with the user emotional experience particularly with respect to empathy of Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar to have curiosity, preconceptions, inspirations, direct experience, listening, or imagination. The combination would allow for, according to the motivation of Chandrasekaran, data to train the machine learning algorithm to learn mapping for association of the user with respect to an appropriate emotional response for the user [0005] particularly with respect to various emotional states of the user [0006], thus better providing a response as a notification to the user, such as appropriately adapting the response to the user and, in particular, with respect to the determined emotional state of the user by modifying data corresponding with notifications in the response to the user [0008].
Regarding claim 13, the claim recites similar limitations as claim 3 – see above.
Claim(s) 4 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Anamandra in view of Olshansky in view of Knoppert in view of Linton in view of Kumar in view of Chandrasekara in view of Hu et al. (US 20220034668, Herein “Hu”).
Regarding claim 4, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar in view of Chandrasekara teaches the limitations of claims 1 and 3, as above.
However, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar in view of Chandrasekar fails to specifically teach The system of claim 3, wherein each of the different machine learning models of the first set of machine learning models are associated with a separate model weighting, and wherein the user interface includes interactive control elements which are configured to receive user inputs modifying the model weightings such that the first tuning matrix can be changed based on different weights applied to the different machine learning models of the first set of machine learning models.
Yet, in a related art, Hu discloses based on a user selection adjusting the weights of the first and second machine learning models [0021].
It would have been obvious to one of ordinary skill in the art prior to the invention’s effective filing date to combine the weight adjustment based on user selection of Hu with the machine learning modeling of Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar in further view of Chandrasekar to have wherein each of the different machine learning models of the first set of machine learning models are associated with a separate model weighting, and wherein the user interface includes interactive control elements which are configured to receive user inputs modifying the model weightings such that the first tuning matrix can be changed based on different weights applied to the different machine learning models of the first set of machine learning models. The combination would allow for, according to the motivation of Hu, a sort of training that can be performed during user operation [0022] such that the user can control the adjustment of the weights of the machine learning models particularly since the training is different based on different users and the contexts [0023].
Regarding claim 14, the claim recites similar limitations as claim 4 – see above.
Claim(s) 6, 7, 8, 16, 17, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Anamandra in view of Olshansky in view of Knoppert in view of Linton in view of Kumar in view of Subramanian (US 20230356556).
Regarding claim 6, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar teaches the limitations of claims 1 and 5, as above.
Furthermore, Anamandra teaches The system of claim 5, wherein the environmental features include at least time, weather, and location of the user (e.g., productivity data such as quantity of productive time, delays in meeting deadlines, etc. and even further situational data such as social situation, family situation, sleep patterns, etc. [0042]).
However, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar fails to specifically teach weather.
Yet, in a related art, Subramanian discloses road conditions [0015].
It would have been obvious to one of ordinary skill in the art prior to the invention’s effective filing date to combine the weather of Subramanian with the conditional modeling of Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar to have weather. The combination would allow for, according to the motivation of Subramanian, analyzing a vehicle’s context such as road conditions and other factors that impact the performance of the vehicle, thus optimizing the driver’s experience for the current driving conditions [0002]; further, for vehicular analysis while the vehicle is driving down the road [0015].
Regarding claim 7, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar in further view of Subramanian teaches the limitations of claims 1, 5 and 6, as above.
Furthermore, Subramanian teaches The system of claim 6, wherein the location of the user further includes a determination of whether the user is currently in transit (determination based on vehicle information such as vehicle velocity, inertial movement, etc. [0004] corresponding with for collecting data while the vehicle is driving down the road [0015]).
Regarding claim 8, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar in further view of Subramanian teaches the limitations of claims 1 and 5-7, as above.
Furthermore, Subramanian teaches The system of claim 7, wherein the determination of whether the user is currently in transit includes obtaining additional features associated with a vehicle in which the user is currently in transit (in addition to velocity, determining inertial movement and road conditions [0004]).
Regarding claim 16, the claim recites similar limitations as claim 6 – see above.
Regarding claim 17, the claim recites similar limitations as claim 7 – see above.
Regarding claim 18, the claim recites similar limitations as claim 8 – see above.
Claim(s) 10 is rejected under 35 U.S.C. 103 as being unpatentable over Anamandra in view of Olshansky in view of Knoppert in view of Linton in view of Kumar in view of Karp et al. (US 20190379615, Herein “Karp”).
Regarding claim 10, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar teaches the limitations of claim 1, as above.
However, Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar fails to specifically teach The system of claim 1, wherein the data set representative of the outcome includes payment interactions associated with the notification.
Yet, in a related art, Karp discloses the prediction system used with notifications sent to the user in anticipation of a user action such as payment [0050]; payment [0035].
It would have been obvious to one of ordinary skill in the art prior to the invention’s effective filing date to combine the outcome includes payment interactions (associated with the notification) of Karp with the notification modeling of Anamandra in view of Olshansky in further view of Knoppert in further view of Linton in further view of Kumar to have wherein the data set representative of the outcome includes payment interactions associated with the notification. The combination would allow for, according to the motivation of Karp, training a prediction model which may be used to anticipate actions of the user based on known user activity [0035] thus enhancing user interaction by providing customers routine actions in an expedited, extended fashion such as by way of payment interactions [0002].
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 DANIEL DUC TRAN whose telephone number is (571)272-6870. The examiner can normally be reached Mon-Fri 8:00-5:00 EST.
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, Viker Lamardo can be reached at (571) 270-5871. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/D.D.T./Examiner, Art Unit 2147
/MICHAEL J HUNTLEY/Supervisory Patent Examiner, Art Unit 2129
Prosecution Insights