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 .
Drawings
The drawings are objected to because in Fig.. Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Specification
The disclosure is objected to because of the following informalities: in paragraphs [.
Appropriate correction is required.
Claim Objections
Claims 1, 8, and 15 ar objected to because of the following informalities: ... the machine-learning model being configured preserve... should be ...the machine-learning model being configured to preserve... Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Independent claims 1 (A method comprising), 8 (A system comprising), and 15(A non-transitory computer-readable medium …) are directed towards a method, a system, and a manufacture respectively. Therefore, these claims, as well as their dependent claims, are directed towards one of the four statutory categories (process, machine, manufacture, or composition of matter).
Claim 1
Step 2A, Prong 1: The claim recites, inter alia:
identifying, based on the media asset, a training dataset associated with the media asset, […]
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to identify information for use in training from a media asset. See MPEP 2106.05(a)(2)(III).
determining that the degree of deviation is less than a threshold;
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to judge that a degree of deviation is less than a given value.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
receiving an identification of a media asset;
This limitation is insignificant extra-solution activity of mere data-gathering. See MPEP 2016.05(g) CyberSource v. Retail Decisions, Inc., 654 F.3d 1366, 1375, 99 USPQ2d 1690, 1694 (Fed. Cir. 2011);
… the training dataset stored in a database that is write protected;
This limitation is recited at a high level of generality and recites use of a generically recited write protected database to store the identified training dataset. Mere instruction that a judicial exception is to be applied using a generically recited write protected database cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
training a machine-learning model using the training dataset, the machine-learning model being configured preserve an integrity of the training dataset by detecting an unauthorized deviation between a feature vector and the training dataset;
This limitation is recited at a high level of generality and recites generic application of training a machine learning model to detect deviations. Mere instruction that a judicial exception is to be applied by a generic training of a machine learning model cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
receiving a request to modify the training dataset, the request including an identification of media usable to modify the training dataset, wherein the media is associated with a characteristic of the media asset;
This limitation is insignificant extra-solution activity of mere data-gathering. See MPEP 2016.05(g);
executing the machine-learning model using a test feature vector derived at least in part from the request, wherein the machine-learning model generates an indication of a degree of deviation between the training dataset and the test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model. Mere instruction that a judicial exception is to be applied using a generic machine-learning model cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
executing a retraining iteration of the machine-learning model in response to determining that the degree of deviation is less than the threshold.
This limitation is recited at a high level of generality and recites generic application of retraining the machine learning model to detect deviations in response to a judgment. Mere instruction that a judicial exception is to be applied by a generic retraining of a machine learning model cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
receiving an identification of a media asset;
MPEP 2106.05(d)(II) buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) indicates that receiving or transmitting data over a network is a well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim).
… the training dataset stored in a database that is write protected;
MPEP 2106.05(d)(II) Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015) indicates that storing and retrieving information in memory is well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim)
training a machine-learning model using the training dataset, the machine-learning model being configured preserve an integrity of the training dataset by detecting an unauthorized deviation between a feature vector and the training dataset;
This limitation is recited at a high level of generality and recites generic application of training a machine learning model to detect deviations. Mere instruction that a judicial exception is to be applied by a generic training of a machine learning model is not significantly more than the judicial exception. See MPEP 2106.05(f);
receiving a request to modify the training dataset, the request including an identification of media usable to modify the training dataset, wherein the media is associated with a characteristic of the media asset;
MPEP 2106.05(d)(II) buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) indicates that receiving or transmitting data over a network is a well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim).
executing the machine-learning model using a test feature vector derived at least in part from the request, wherein the machine-learning model generates an indication of a degree of deviation between the training dataset and the test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model to apply a request. Mere instruction that a judicial exception is to be applied using a generic machine-learning model is not significantly more than the judicial exception. See MPEP 2106.05(f);
executing a retraining iteration of the machine-learning model in response to determining that the degree of deviation is less than the threshold.
This limitation is recited at a high level of generality and recites generic application of retraining the machine learning model to detect deviations in response to a judgment. Mere instruction that a judicial exception is to be applied by a generic retraining of a machine learning model is not significantly more than the judicial exception. See MPEP 2106.05(f);
Claim 2
Step 2A, Prong 1: This claim does not recite any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
wherein the training dataset includes a set of media that represents a canon of the media asset.
This limitation recites that the identified training dataset represents a canon of the media asset which is merely indicating a field of use for the identified training dataset. See MPEP 2106.05(h)
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the training dataset includes a set of media that represents a canon of the media asset.
This limitation recites limiting that the identified training dataset includes a set of media that represents a canon of the media asset which is merely indicating a field of use for the identified training dataset. See MPEP 2106.05(h)
Claim 3
Step 2A, Prong 1: The claim recites, inter alia:
determining the new degree of deviation between the training dataset and the subsequent request is greater than the threshold;
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to evaluate a degree of deviation between the identified training dataset and a new request and judge if it is greater than a given value.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
receiving a subsequent request to modify the training dataset;
This limitation is insignificant extra-solution activity of mere data-gathering. See MPEP 2016.05(g);
executing the machine-learning model using a new test feature vector derived from the subsequent request, wherein the machine-learning model generates a new degree of deviation between the training dataset and the new test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model to apply a request. Mere instruction that a judicial exception is to be applied using a generic machine-learning model does not integrate the judicial exception into a practical application. See MPEP 2106.05(f);
preventing the training dataset from being modified by removing the subsequent request.
This limitation is recited at a high level of generality and recites generic removal of a request. Mere instruction that a judicial exception is not to be applied does not integrate the judicial exception into a practical application. See MPEP 2106.05(f);
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
receiving a subsequent request to modify the training dataset;
MPEP 2106.05(d)(II) buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) indicates that receiving or transmitting data over a network is a well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim).
executing the machine-learning model using a new test feature vector derived from the subsequent request, wherein the machine-learning model generates a new degree of deviation between the training dataset and the new test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model to apply a request. Mere instruction that a judicial exception is to be applied using a generic machine-learning model is not significantly more than a judicial exception. See MPEP 2106.05(f);
preventing the training dataset from being modified by removing the subsequent request.
This limitation is recited at a high level of generality and recites generic removal of a request. Mere instruction that a judicial exception is not to be applied is not significantly more than a judicial exception. See MPEP 2106.05(f);
Claim 4
Step 2A, Prong 1: This claim does not recite any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
wherein the machine-learning model is a large language model.
This limitation is recited at a high level of generality and recites use of a generic large language model. Mere instruction that a judicial exception is to be applied using a generic large language model does not integrate the judicial exception into a practical application. See MPEP 2106.05(f);
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the machine-learning model is a large language model.
This limitation is recited at a high level of generality and recites use of a generic large language model. Mere instruction that a judicial exception is to be applied using a generic large language model is not significantly more than a judicial exception. See MPEP 2106.05(f);
Claim 5
Step 2A, Prong 1: This claim does not recite any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
wherein the characteristic of the media asset corresponds to a character or book title.
This limitation recites limiting the characteristic of the media asset to the particular field of use of a character or book title. Mere indication to limit an abstract idea to a field of use does not integrate the judicial exception into a practical application. See MPEP 2106.05(h)
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the characteristic of the media asset corresponds to a character or book title.
This limitation recites limiting the characteristic of the media asset to the particular field of use of a character or book title. Mere indication to limit an abstract idea to a field of use is not significantly more than a judicial exception. See MPEP 2106.05(h)
Claim 6
Step 2A, Prong 1: This claim does not have any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
wherein the media includes one or more strings, an image, or a video segment.
This limitation recites limiting the media to include the particular field of use of strings, images, or video segments. Mere indication to limit an abstract idea to include a field of use does not integrate the judicial exception into a practical application. See MPEP 2106.05(h)
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
wherein the media includes one or more strings, an image, or a video segment.
This limitation recites limiting the media to include the particular field of use of strings, images, or video segments. Mere indication to limit an abstract idea to include a field of use is not significantly more than a judicial exception. See MPEP 2106.05(h)
Claim 7
Step 2A, Prong 1: The claim recites, inter alia:
wherein identifying the training dataset includes defining the training dataset from data associated with the media asset and data associated with a second media asset, wherein the characteristic is included in both the data associated with the media asset and data associated with the second media asset.
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to identify the training data using data associated with two media assets that share a characteristic.
Step 2A, Prong 2: This claim does not have any further additional elements.
Step 2B: This claim does not have any further additional elements.
Claim 8
Step 2A, Prong 1: The claim recites, inter alia:
identifying, based on the media asset, a training dataset associated with the media asset, […]
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to identify information for use in training from a media asset
determining that the degree of deviation is less than a threshold;
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to judge that a degree of deviation is less than a given value.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
A system comprising: one or more processors;
This limitation is recited at a high level of generality and recites use of generic computer equipment to apply the abstract idea. Mere instruction that a judicial exception is to be applied using generic computer equipment cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
a non-transitory computer-readable medium storing instructions that when executed by the one or more processors, cause the one or more processors to perform operations including:
This limitation is recited at a high level of generality and recites use of generic computer equipment to apply the abstract idea. Mere instruction that a judicial exception is to be applied using generic computer equipment cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
receiving an identification of a media asset;
This limitation is insignificant extra-solution activity of mere data-gathering. See MPEP 2016.05(g);
… the training dataset stored in a database that is write protected;
This limitation is recited at a high level of generality and recites use of a generically recited write protected database to store the identified training dataset. Mere instruction that a judicial exception is to be applied using a generically recited write protected database cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
training a machine-learning model using the training dataset, the machine-learning model being configured preserve an integrity of the training dataset by detecting an unauthorized deviation between a feature vector and the training dataset;
This limitation is recited at a high level of generality and recites generic application of training a machine learning model to detect deviations. Mere instruction that a judicial exception is to be applied by a generic training of a machine learning model cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
receiving a request to modify the training dataset, the request including an identification of media usable to modify the training dataset, wherein the media is associated with a characteristic of the media asset;
This limitation is insignificant extra-solution activity of mere data-gathering. See MPEP 2016.05(g);
executing the machine-learning model using a test feature vector derived at least in part from the request, wherein the machine-learning model generates an indication of a degree of deviation between the training dataset and the test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model. Mere instruction that a judicial exception is to be applied using a generic machine-learning model cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
executing a retraining iteration of the machine-learning model in response to determining that the degree of deviation is less than the threshold.
This limitation is recited at a high level of generality and recites generic application of retraining the machine learning model to detect deviations in response to a judgment. Mere instruction that a judicial exception is to be applied by a generic retraining of a machine learning model cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
A system comprising: one or more processors;
This limitation is recited at a high level of generality and recites use of generic computer equipment to apply the abstract idea. Mere instruction that a judicial exception is to be applied using generic computer equipment is not significantly more than the judicial exception. See MPEP 2106.05(f);
a non-transitory computer-readable medium storing instructions that when executed by the one or more processors, cause the one or more processors to perform operations including:
This limitation is recited at a high level of generality and recites use of generic computer equipment to apply the abstract idea. Mere instruction that a judicial exception is to be applied using generic computer equipment is not significantly more than the judicial exception. See MPEP 2106.05(f);
receiving an identification of a media asset;
MPEP 2106.05(d)(II) buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) indicates that receiving or transmitting data over a network is a well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim).
… the training dataset stored in a database that is write protected;
This limitation is recited at a high level of generality and recites use of a generically recited write protected database to store the identified training dataset. Mere instruction that a judicial exception is to be applied using a generically recited write protected database is not significantly more than the judicial exception. See MPEP 2106.05(f);
training a machine-learning model using the training dataset, the machine-learning model being configured preserve an integrity of the training dataset by detecting an unauthorized deviation between a feature vector and the training dataset;
This limitation is recited at a high level of generality and recites generic application of training a machine learning model to detect deviations. Mere instruction that a judicial exception is to be applied by a generic training of a machine learning model is not significantly more than the judicial exception. See MPEP 2106.05(f);
receiving a request to modify the training dataset, the request including an identification of media usable to modify the training dataset, wherein the media is associated with a characteristic of the media asset;
MPEP 2106.05(d)(II) buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) indicates that receiving or transmitting data over a network is a well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim).
executing the machine-learning model using a test feature vector derived at least in part from the request, wherein the machine-learning model generates an indication of a degree of deviation between the training dataset and the test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model to apply a request. Mere instruction that a judicial exception is to be applied using a generic machine-learning model is not significantly more than the judicial exception. See MPEP 2106.05(f);
executing a retraining iteration of the machine-learning model in response to determining that the degree of deviation is less than the threshold.
This limitation is recited at a high level of generality and recites generic application of retraining the machine learning model to detect deviations in response to a judgment. Mere instruction that a judicial exception is to be applied by a generic retraining of a machine learning model is not significantly more than the judicial exception. See MPEP 2106.05(f);
Claim 9
Step 2A, Prong 1: This claim does not recite any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
wherein the training dataset includes a set of media that represents a canon of the media asset.
This limitation recites that the identified training dataset represents a canon of the media asset which is merely indicating a field of use for the identified training dataset. See MPEP 2106.05(h)
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the training dataset includes a set of media that represents a canon of the media asset.
This limitation recites limiting that the identified training dataset includes a set of media that represents a canon of the media asset which is merely indicating a field of use for the identified training dataset. See MPEP 2106.05(h)
Claim 10
Step 2A, Prong 1: The claim recites, inter alia:
determining the new degree of deviation between the training dataset and the subsequent request is greater than the threshold;
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to evaluate a degree of deviation between the identified training dataset and a new request and judge if it is greater than a given value.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
receiving a subsequent request to modify the training dataset;
This limitation is insignificant extra-solution activity of mere data-gathering. See MPEP 2016.05(g);
executing the machine-learning model using a new test feature vector derived from the subsequent request, wherein the machine-learning model generates a new degree of deviation between the training dataset and the new test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model to apply a request. Mere instruction that a judicial exception is to be applied using a generic machine-learning model does not integrate the judicial exception into a practical application. See MPEP 2106.05(f);
preventing the training dataset from being modified by removing the subsequent request.
This limitation is recited at a high level of generality and recites generic removal of a request. Mere instruction that a judicial exception is not to be applied does not integrate the judicial exception into a practical application. See MPEP 2106.05(f);
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
receiving a subsequent request to modify the training dataset;
MPEP 2106.05(d)(II) buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) indicates that receiving or transmitting data over a network is a well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim).
executing the machine-learning model using a new test feature vector derived from the subsequent request, wherein the machine-learning model generates a new degree of deviation between the training dataset and the new test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model to apply a request. Mere instruction that a judicial exception is to be applied using a generic machine-learning model is not significantly more than a judicial exception. See MPEP 2106.05(f);
preventing the training dataset from being modified by removing the subsequent request.
This limitation is recited at a high level of generality and recites generic removal of a request. Mere instruction that a judicial exception is not to be applied is not significantly more than a judicial exception. See MPEP 2106.05(f);
Claim 11
Step 2A, Prong 1: This claim does not recite any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
wherein the machine-learning model is a large language model.
This limitation is recited at a high level of generality and recites use of a generic large language model. Mere instruction that a judicial exception is to be applied using a generic large language model does not integrate the judicial exception into a practical application. See MPEP 2106.05(f);
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the machine-learning model is a large language model.
This limitation is recited at a high level of generality and recites use of a generic large language model. Mere instruction that a judicial exception is to be applied using a generic large language model is not significantly more than a judicial exception. See MPEP 2106.05(f);
Claim 12
Step 2A, Prong 1: This claim does not recite any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
wherein the characteristic of the media asset corresponds to a character or book title.
This limitation recites limiting the characteristic of the media asset to the particular field of use of a character or book title. Mere indication to limit an abstract idea to a field of use does not integrate the judicial exception into a practical application. See MPEP 2106.05(h)
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the characteristic of the media asset corresponds to a character or book title.
This limitation recites limiting the characteristic of the media asset to the particular field of use of a character or book title. Mere indication to limit an abstract idea to a field of use is not significantly more than a judicial exception. See MPEP 2106.05(h)
Claim 13
Step 2A, Prong 1: This claim does not have any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
wherein the media includes one or more strings, an image, or a video segment.
This limitation recites limiting the media to include the particular field of use of strings, images, or video segments. Mere indication to limit an abstract idea to include a field of use does not integrate the judicial exception into a practical application. See MPEP 2106.05(h)
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
wherein the media includes one or more strings, an image, or a video segment.
This limitation recites limiting the media to include the particular field of use of strings, images, or video segments. Mere indication to limit an abstract idea to include a field of use is not significantly more than a judicial exception. See MPEP 2106.05(h)
Claim 14
Step 2A, Prong 1: The claim recites, inter alia:
wherein identifying the training dataset includes defining the training dataset from data associated with the media asset and data associated with a second media asset, wherein the characteristic is included in both the data associated with the media asset and data associated with the second media asset.
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to identify the training data using data associated with two media assets that share a characteristic.
Step 2A, Prong 2: This claim does not have any further additional elements.
Step 2B: This claim does not have any further additional elements.
Claim 15
Step 2A, Prong 1: The claim recites, inter alia:
identifying, based on the media asset, a training dataset associated with the media asset, […]
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to identify information for use in training from a media asset
determining that the degree of deviation is less than a threshold;
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to judge that a degree of deviation is less than a given value.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
A non-transitory computer-readable medium storing instructions that when executed by one or more processors, cause the one or more processors to perform operations including:
This limitation is recited at a high level of generality and recites use of generic computer equipment to apply the abstract idea. Mere instruction that a judicial exception is to be applied using generic computer equipment cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
receiving an identification of a media asset;
This limitation is insignificant extra-solution activity of mere data-gathering. See MPEP 2016.05(g);
… the training dataset stored in a database that is write protected;
This limitation is recited at a high level of generality and recites use of a generically recited write protected database to store the identified training dataset. Mere instruction that a judicial exception is to be applied using a generically recited write protected database cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
training a machine-learning model using the training dataset, the machine-learning model being configured preserve an integrity of the training dataset by detecting an unauthorized deviation between a feature vector and the training dataset;
This limitation is recited at a high level of generality and recites generic application of training a machine learning model to detect deviations. Mere instruction that a judicial exception is to be applied by a generic training of a machine learning model cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
receiving a request to modify the training dataset, the request including an identification of media usable to modify the training dataset, wherein the media is associated with a characteristic of the media asset;
This limitation is insignificant extra-solution activity of mere data-gathering. See MPEP 2016.05(g);
executing the machine-learning model using a test feature vector derived at least in part from the request, wherein the machine-learning model generates an indication of a degree of deviation between the training dataset and the test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model. Mere instruction that a judicial exception is to be applied using a generic machine-learning model cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
executing a retraining iteration of the machine-learning model in response to determining that the degree of deviation is less than the threshold.
This limitation is recited at a high level of generality and recites generic application of retraining the machine learning model to detect deviations in response to a judgment. Mere instruction that a judicial exception is to be applied by a generic retraining of a machine learning model cannot integrate the judicial exception into a practical application. See MPEP 2106.05(f);
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
A non-transitory computer-readable medium storing instructions that when executed by one or more processors, cause the one or more processors to perform operations including:
This limitation is recited at a high level of generality and recites use of generic computer equipment to apply the abstract idea. Mere instruction that a judicial exception is to be applied using generic computer equipment is not significantly more than the judicial exception. See MPEP 2106.05(f);
receiving an identification of a media asset;
MPEP 2106.05(d)(II) buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) indicates that receiving or transmitting data over a network is a well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim).
… the training dataset stored in a database that is write protected;
This limitation is recited at a high level of generality and recites use of a generically recited write protected database to store the identified training dataset. Mere instruction that a judicial exception is to be applied using a generically recited write protected database is not significantly more than the judicial exception. See MPEP 2106.05(f);
training a machine-learning model using the training dataset, the machine-learning model being configured preserve an integrity of the training dataset by detecting an unauthorized deviation between a feature vector and the training dataset;
This limitation is recited at a high level of generality and recites generic application of training a machine learning model to detect deviations. Mere instruction that a judicial exception is to be applied by a generic training of a machine learning model is not significantly more than the judicial exception. See MPEP 2106.05(f);
receiving a request to modify the training dataset, the request including an identification of media usable to modify the training dataset, wherein the media is associated with a characteristic of the media asset;
MPEP 2106.05(d)(II) buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) indicates that receiving or transmitting data over a network is a well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim).
executing the machine-learning model using a test feature vector derived at least in part from the request, wherein the machine-learning model generates an indication of a degree of deviation between the training dataset and the test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model to apply a request. Mere instruction that a judicial exception is to be applied using a generic machine-learning model is not significantly more than the judicial exception. See MPEP 2106.05(f);
executing a retraining iteration of the machine-learning model in response to determining that the degree of deviation is less than the threshold.
This limitation is recited at a high level of generality and recites generic application of retraining the machine learning model to detect deviations in response to a judgment. Mere instruction that a judicial exception is to be applied by a generic retraining of a machine learning model is not significantly more than the judicial exception. See MPEP 2106.05(f);
Claim 16
Step 2A, Prong 1: This claim does not recite any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
wherein the training dataset includes a set of media that represents a canon of the media asset.
This limitation recites that the identified training dataset represents a canon of the media asset which is merely indicating a field of use for the identified training dataset. See MPEP 2106.05(h)
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the training dataset includes a set of media that represents a canon of the media asset.
This limitation recites limiting that the identified training dataset includes a set of media that represents a canon of the media asset which is merely indicating a field of use for the identified training dataset. See MPEP 2106.05(h)
Claim 17
Step 2A, Prong 1: The claim recites, inter alia:
determining the new degree of deviation between the training dataset and the subsequent request is greater than the threshold;
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to evaluate a degree of deviation between the identified training dataset and a new request and judge if it is greater than a given value.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
receiving a subsequent request to modify the training dataset;
This limitation is insignificant extra-solution activity of mere data-gathering. See MPEP 2016.05(g);
executing the machine-learning model using a new test feature vector derived from the subsequent request, wherein the machine-learning model generates a new degree of deviation between the training dataset and the new test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model to apply a request. Mere instruction that a judicial exception is to be applied using a generic machine-learning model does not integrate the judicial exception into a practical application. See MPEP 2106.05(f);
preventing the training dataset from being modified by removing the subsequent request.
This limitation is recited at a high level of generality and recites generic removal of a request. Mere instruction that a judicial exception is not to be applied does not integrate the judicial exception into a practical application. See MPEP 2106.05(f);
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
receiving a subsequent request to modify the training dataset;
MPEP 2106.05(d)(II) buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355, 112 USPQ2d 1093, 1096 (Fed. Cir. 2014) indicates that receiving or transmitting data over a network is a well-understood, routine and conventional activity when recited in a merely generic manner (as it is in the present claim).
executing the machine-learning model using a new test feature vector derived from the subsequent request, wherein the machine-learning model generates a new degree of deviation between the training dataset and the new test feature vector;
This limitation is recited at a high level of generality and recites generic use of the machine learning model to apply a request. Mere instruction that a judicial exception is to be applied using a generic machine-learning model is not significantly more than a judicial exception. See MPEP 2106.05(f);
preventing the training dataset from being modified by removing the subsequent request.
This limitation is recited at a high level of generality and recites generic removal of a request. Mere instruction that a judicial exception is not to be applied is not significantly more than a judicial exception. See MPEP 2106.05(f);
Claim 18
Step 2A, Prong 1: This claim does not recite any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
wherein the machine-learning model is a large language model.
This limitation is recited at a high level of generality and recites use of a generic large language model. Mere instruction that a judicial exception is to be applied using a generic large language model does not integrate the judicial exception into a practical application. See MPEP 2106.05(f);
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the machine-learning model is a large language model.
This limitation is recited at a high level of generality and recites use of a generic large language model. Mere instruction that a judicial exception is to be applied using a generic large language model is not significantly more than a judicial exception. See MPEP 2106.05(f);
Claim 19
Step 2A, Prong 1: This claim does not recite any further abstract ideas.
Step 2A, Prong 2: The additional elements recited in this claim do not integrate the judicial exception into a practical application.
Additional Elements:
wherein the characteristic of the media asset corresponds to a character or book title.
This limitation recites limiting the characteristic of the media asset to the particular field of use of a character or book title. Mere indication to limit an abstract idea to a field of use does not integrate the judicial exception into a practical application. See MPEP 2106.05(h)
Step 2B: This claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional Elements:
wherein the characteristic of the media asset corresponds to a character or book title.
This limitation recites limiting the characteristic of the media asset to the particular field of use of a character or book title. Mere indication to limit an abstract idea to a field of use is not significantly more than a judicial exception. See MPEP 2106.05(h)
Claim 20
Step 2A, Prong 1: The claim recites, inter alia:
wherein identifying the training dataset includes defining the training dataset from data associated with the media asset and data associated with a second media asset, wherein the characteristic is included in both the data associated with the media asset and data associated with the second media asset.
This limitation recites a mental process using observation, evaluation, judgment, and opinion, with aid of pen and paper, to identify the training data using data associated with two media assets that share a characteristic.
Step 2A, Prong 2: This claim does not have any further additional elements.
Step 2B: This claim does not have any further additional elements.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over US 20240221302 A1 by Gelfenbeyn et al., hereafter Gelfenbeyn, in view of US 20250037008 A1 by Mara et al., hereafter Mara.
Regarding claim 1, Gelfenbeyn teaches:
A method comprising:
receiving an identification of a media asset;
identifying, based on the media asset (“canonical source information”), a training dataset associated with the media asset (“reference knowledge store”) […]; ((Gelfenbeyn) Paragraph [0098], “The reference knowledge store may be associated with a canonical source information related to one of the following: a known fictional character, a known fictional world, a known environment, a known person, and so forth.”)
training a machine-learning model using the training dataset ((Gelfenbeyn) Paragraph [0026], “The platform may utilize a common knowledge to train the AI character model in order to interact with the users.”), the machine-learning model being configured [to] preserve an integrity of the training dataset by detecting an unauthorized deviation (“determine difference”) between a feature vector (“knowledge store”) and the training dataset (“reference knowledge store”); ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, defining the scope of knowledge may include comparing the knowledge store to a reference knowledge store to determine differences between the knowledge store and the reference knowledge store.” Determining differences is detecting an unauthorized deviation.)
receiving a request to modify the training dataset, the request including an identification of media usable to modify the training dataset, wherein the media is associated with a characteristic of the media asset; ((Gelfenbeyn) Paragraph [0077], “An identity profile 718 selected by the user may specify elements of an AI character (e.g., role, interests) which may have an influence on how the AI character pursues goals” A user selecting an identity profile is the user sending a request to add to the knowledge store, which is modifying the training dataset. An AI character is a media asset of a media usable to modify the training dataset, and the elements of the AI character are characteristics of the media asset.)
executing the machine-learning model using a test feature vector derived at least in part from the request, wherein the machine-learning model generates an indication of a degree of deviation between the training dataset and the test feature vector; ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, defining the scope of knowledge may include comparing the knowledge store to a reference knowledge store to determine differences between the knowledge store and the reference knowledge store.” The knowledge store is a test feature vector derived from the identity profile(request) compared to the reference knowledge store which is the training dataset. Determining differences is generating an indication of a degree of deviation.)
Gelfenbeyn does not teach:
… the training dataset stored in a database that is write protected;
determining that the degree of deviation is less than a threshold; and
executing a retraining iteration of the machine-learning model in response to determining that the degree of deviation is less than the threshold.
Mara teaches:
A dataset being stored in a database that is write protected; ((Mara) Paragraph [0127], “for instance, and without limitation, temporally sequential listing may be readable and modifiable publicly, may be publicly readable but writable only by entities and/or devices having access privileges established by password protection, confidence level, or any device authentication procedure or facilities described herein, or may be readable and/or writable only by entities and/or devices having such access privileges” A dataset being stored in a database where only entities having certain access privileges to write to is being stored in a write protected database; Paragraph [0087], “In some cases, authentication datum may serve as a key to protect plurality of provider models 116 from unauthorized access and maintaining security and privacy.”)
Determining that a degree of deviation is not greater than a threshold; and ((Mara) Paragraph [0142], “‘Sanitizing’ training data, as used in this disclosure, is a process whereby training examples are removed that interfere with convergence of a machine-learning model and/or process to a useful result…a value that is more than a threshold number of standard deviations away from an average, mean, or expected value, for instance, may be eliminated.” A value that is not eliminated has been determined to not be more than a threshold degree of deviation away from a compared value.)
Executing a retraining iteration of the machine-learning model in response to accepting added training data input when the degree of deviation is not greater than the threshold. ((Mara) Paragraph [0155], “any process of training, retraining, deployment, and/or instantiation of any machine-learning model and/or algorithm may be performed and/or repeated after an initial deployment and/or instantiation to correct, refine, and/or improve the machine-learning model and/or algorithm … Event-based retraining, deployment, and/or instantiation may alternatively or additionally be triggered by receipt and/or generation of one or more new training examples” Receipt of new training examples is accepting added training data input which happens when a value is not more than a threshold number of standard deviations away from a compared value;)
Mara and Gelfenbeyn are analogous art because they are in the same area of invention: communicating with users using artificial intelligence.
Thus, it would have been obvious to a person having ordinary skill in the art before the effective filing date, having the references in front of them, to have combined storing a dataset in a write-protected database, as Mara teaches, with the identifying a training dataset from a media asset, as taught by Gelfenbeyn, and to have combined determining a degree of deviation is not greater than a threshold and executing a retraining iteration as a response to events resulting from the determination, as Mara teaches, with determining a degree of deviation between a knowledge store and a reference knowledge store, as taught by Gelfenbeyn. The application of the technique of storing the identified training dataset in a write-protected database would have been motivated to protect models from unauthorized access and to maintain security and privacy. The application of the determination that the degree of deviation is not greater than a threshold would have been motivated to check for inputs that would interfere with convergence to a useful result and executing a retraining iteration as a response to events resulting from the determination would have been motivated to refine the machine-learning model. These applications of known techniques from the teachings of Mara into the method taught by Gelfenbeyn would have yielded the predictable result that is disclosed in claim 1 of the instant application. The Examiner notes that this motivation applies to all dependent and/or otherwise subsequently addressed claims.
Regarding claim 2, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 1, and additionally, Gelfenbeyn teaches:
wherein the training dataset includes a set of media that represents a canon of the media asset. ((Gelfenbeyn) Paragraph [0098], “The reference knowledge store may be associated with a canonical source information related to one of the following: a known fictional character, a known fictional world, a known environment, a known person, and so forth.” A reference knowledge store is a training dataset.)
Regarding claim 3, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 1, and additionally, Gelfenbeyn teaches:
receiving a subsequent request to modify the training dataset; ((Gelfenbeyn) Paragraph [0077], “An identity profile 718 selected by the user may specify elements of an AI character (e.g., role, interests) which may have an influence on how the AI character pursues goals” A user selecting an identity profile is the user sending a request to add to the knowledge store, which is modifying the training dataset. An AI character is a media asset of a media usable to modify the training dataset, and the elements of the AI character are characteristics of the media asset.)
executing the machine-learning model using a new test feature vector derived from the subsequent request, wherein the machine-learning model generates a new degree of deviation between the training dataset and the new test feature vector; ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, defining the scope of knowledge may include comparing the knowledge store to a reference knowledge store to determine differences between the knowledge store and the reference knowledge store.” The knowledge store is a test feature vector derived from the identity profile(request) compared to the reference knowledge store which is the training dataset. Determining differences is generating an indication of a degree of deviation.)
Gelfenbeyn does not teach:
determining the new degree of deviation between the training dataset and the subsequent request is greater than the threshold; and
preventing the training dataset from being modified by removing the subsequent request.
Mara teaches:
Determining the degree of deviation is greater than a threshold; and
Preventing the training dataset from being modified by removing the input. ((Mara) Paragraph [0142], “‘Sanitizing’ training data, as used in this disclosure, is a process whereby training examples are removed that interfere with convergence of a machine-learning model and/or process to a useful result…a value that is more than a threshold number of standard deviations away from an average, mean, or expected value, for instance, may be eliminated.”)
Mara and Gelfenbeyn are analogous art because they are in the same area of invention: communicating with users using artificial intelligence.
Thus, it would have been obvious to a person having ordinary skill in the art before the effective filing date, having the references in front of them, to have combined determining a degree of deviation is more than a threshold and eliminating the input in response to the determination, as Mara teaches, with determining a degree of deviation between a knowledge store and a reference knowledge store, as taught by Gelfenbeyn. The application of the determination that the degree of deviation is more than a threshold and eliminating the input in response to the determination would have been motivated to remove inputs that would interfere with convergence to a useful result. This application of a known technique from the teachings of Mara into the method taught by Gelfenbeyn would have yielded the predictable results that is disclosed in claim 3 of the instant application.
Regarding claim 4, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 1, and additionally, Gelfenbeyn teaches:
wherein the machine-learning model is a large language model. ((Gelfenbeyn) Paragraph [0013], “FIG. 4 is an architecture diagram that shows using a surrounding architecture of an AI character model to control an output and behavior generated by a large language model (LLM), according to an example embodiment.”)
Regarding claim 5, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 1, and additionally, Gelfenbeyn teaches:
wherein the characteristic of the media asset corresponds to a character or book title. ((Gelfenbeyn) Paragraph [0026], “In one example embodiment, the platform may receive a description of a character and generate an AI character model capable of interacting with users verbally and through emotions, gestures, actions, and movements.”)
Regarding claim 6, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 1, and additionally, Gelfenbeyn teaches:
wherein the media includes one or more strings, an image, or a video segment. ((Gelfenbeyn) Paragraph [0098], “For example, defining the scope of knowledge may include selecting, for the AI character, a canonical source or corpus of information (e.g., all Star Wars® canon) as the source of truth.” All Star Wars® canon includes books, which include strings, and movies, which include video segments.)
Regarding claim 7, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 1, and additionally, Gelfenbeyn teaches:
wherein identifying the training dataset includes defining the training dataset from data associated with the media asset and data associated with a second media asset, wherein the characteristic is included in both the data associated with the media asset and data associated with the second media asset. ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, the canonical source information may include official information, generally known information, information from a predefined source, information provided by authors of virtual characters or virtual worlds, and so forth.” Different sources of information are separate media assets. Separate media assets concerning the same canon may include the same characters.)
Regarding claim 8, Gelfenbeyn teaches:
A system comprising:
one or more processors; ((Gelfenbeyn) Paragraph [0110], “The computer system 1100 may include one or more processor(s) 1102, a memory 1104…”)
a non-transitory computer-readable medium storing instructions that when executed by the one or more processors, cause the one or more processors to perform operations including: ((Gelfenbeyn) Paragraph [0111], “The memory 1104, in some example embodiments, may refer to a non-transitory computer-readable storage medium or a computer-readable storage device… the memory 1104 is used to store program instructions for execution by the processor(s)”)
receiving an identification of a media asset;
identifying, based on the media asset, a training dataset associated with the media asset […]; ((Gelfenbeyn) Paragraph [0098], “The reference knowledge store may be associated with a canonical source information related to one of the following: a known fictional character, a known fictional world, a known environment, a known person, and so forth.” A reference knowledge store is a training dataset. A canonical source information is a media asset.)
training a machine-learning model using the training dataset ((Gelfenbeyn) Paragraph [0026], “The platform may utilize a common knowledge to train the AI character model in order to interact with the users.”), the machine-learning model being configured [to] preserve an integrity of the training dataset by detecting an unauthorized deviation between a feature vector and the training dataset; ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, defining the scope of knowledge may include comparing the knowledge store to a reference knowledge store to determine differences between the knowledge store and the reference knowledge store.” The knowledge store is a feature vector compared to the reference knowledge store which is the training dataset. Determining differences is detecting an unauthorized deviation.)
receiving a request to modify the training dataset, the request including an identification of media usable to modify the training dataset, wherein the media is associated with a characteristic of the media asset; ((Gelfenbeyn) Paragraph [0077], “An identity profile 718 selected by the user may specify elements of an AI character (e.g., role, interests) which may have an influence on how the AI character pursues goals” A user selecting an identity profile is the user sending a request to add to the knowledge store, which is modifying the training dataset. An AI character is a media asset of a media usable to modify the training dataset, and the elements of the AI character are characteristics of the media asset.)
executing the machine-learning model using a test feature vector derived at least in part from the request, wherein the machine-learning model generates an indication of a degree of deviation between the training dataset and the test feature vector; ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, defining the scope of knowledge may include comparing the knowledge store to a reference knowledge store to determine differences between the knowledge store and the reference knowledge store.” The knowledge store is a test feature vector derived from the identity profile(request) compared to the reference knowledge store which is the training dataset. Determining differences is generating an indication of a degree of deviation.)
Gelfenbeyn does not teach:
… the training dataset stored in a database that is write protected;
determining that the degree of deviation is less than a threshold; and
executing a retraining iteration of the machine-learning model in response to determining that the degree of deviation is less than the threshold.
Mara teaches:
A dataset being stored in a database that is write protected; ((Mara) Paragraph [0127], “for instance, and without limitation, temporally sequential listing may be readable and modifiable publicly, may be publicly readable but writable only by entities and/or devices having access privileges established by password protection, confidence level, or any device authentication procedure or facilities described herein, or may be readable and/or writable only by entities and/or devices having such access privileges” A dataset being stored in a database where only entities having certain access privileges to write to is being stored in a write protected database; Paragraph [0087], “In some cases, authentication datum may serve as a key to protect plurality of provider models 116 from unauthorized access and maintaining security and privacy.”)
Determining that a degree of deviation is not greater than a threshold; and ((Mara) Paragraph [0142], “‘Sanitizing’ training data, as used in this disclosure, is a process whereby training examples are removed that interfere with convergence of a machine-learning model and/or process to a useful result…a value that is more than a threshold number of standard deviations away from an average, mean, or expected value, for instance, may be eliminated.” A value that is not eliminated has been determined to not be more than a threshold degree of deviation away from a compared value.)
Executing a retraining iteration of the machine-learning model in response to accepting added training data input when the degree of deviation is not greater than the threshold. ((Mara) Paragraph [0155], “any process of training, retraining, deployment, and/or instantiation of any machine-learning model and/or algorithm may be performed and/or repeated after an initial deployment and/or instantiation to correct, refine, and/or improve the machine-learning model and/or algorithm … Event-based retraining, deployment, and/or instantiation may alternatively or additionally be triggered by receipt and/or generation of one or more new training examples” Receipt of new training examples is accepting added training data input which happens when a value is not more than a threshold number of standard deviations away from a compared value;)
Mara and Gelfenbeyn are analogous art because they are in the same area of invention: communicating with users using artificial intelligence.
Thus, it would have been obvious to a person having ordinary skill in the art before the effective filing date, having the references in front of them, to have combined storing a dataset in a write-protected database, as Mara teaches, with the identifying a training dataset from a media asset, as taught by Gelfenbeyn, and to have combined determining a degree of deviation is not greater than a threshold and executing a retraining iteration as a response to events resulting from the determination, as Mara teaches, with determining a degree of deviation between a knowledge store and a reference knowledge store, as taught by Gelfenbeyn. The application of the technique of storing the identified training dataset in a write-protected database would have been motivated to protect models from unauthorized access and to maintain security and privacy. The application of the determination that the degree of deviation is not greater than a threshold would have been motivated to check for inputs that would interfere with convergence to a useful result and executing a retraining iteration as a response to events resulting from the determination would have been motivated to refine the machine-learning model. These applications of known techniques from the teachings of Mara into the instructions taught by Gelfenbeyn would have yielded the predictable result that is disclosed in claim 8 of the instant application. The Examiner notes that this motivation applies to all dependent and/or otherwise subsequently addressed claims.
Regarding claim 9, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 8, and additionally, Gelfenbeyn teaches:
wherein the training dataset includes a set of media that represents a canon of the media asset. ((Gelfenbeyn) Paragraph [0098], “The reference knowledge store may be associated with a canonical source information related to one of the following: a known fictional character, a known fictional world, a known environment, a known person, and so forth.” A reference knowledge store is a training dataset.)
Regarding claim 10, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 8, and additionally, Gelfenbeyn teaches:
wherein the operations further include:
receiving a subsequent request to modify the training dataset; ((Gelfenbeyn) Paragraph [0077], “An identity profile 718 selected by the user may specify elements of an AI character (e.g., role, interests) which may have an influence on how the AI character pursues goals” A user selecting an identity profile is the user sending a request to add to the knowledge store, which is modifying the training dataset. An AI character is a media asset of a media usable to modify the training dataset, and the elements of the AI character are characteristics of the media asset.)
executing the machine-learning model using a new test feature vector derived from the subsequent request, wherein the machine-learning model generates a new degree of deviation between the training dataset and the new test feature vector; ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, defining the scope of knowledge may include comparing the knowledge store to a reference knowledge store to determine differences between the knowledge store and the reference knowledge store.” The knowledge store is a test feature vector derived from the identity profile(request) compared to the reference knowledge store which is the training dataset. Determining differences is generating an indication of a degree of deviation.)
Gelfenbeyn does not teach:
determining the new degree of deviation between the training dataset and the subsequent request is greater than the threshold; and
preventing the training dataset from being modified by removing the subsequent request.
Mara teaches:
Determining the degree of deviation is greater than a threshold; and
Preventing the training dataset from being modified by removing the input. ((Mara) Paragraph [0142], “‘Sanitizing’ training data, as used in this disclosure, is a process whereby training examples are removed that interfere with convergence of a machine-learning model and/or process to a useful result…a value that is more than a threshold number of standard deviations away from an average, mean, or expected value, for instance, may be eliminated.”)
Mara and Gelfenbeyn are analogous art because they are in the same area of invention: communicating with users using artificial intelligence.
Thus, it would have been obvious to a person having ordinary skill in the art before the effective filing date, having the references in front of them, to have combined determining a degree of deviation is more than a threshold and eliminating the input in response to the determination, as Mara teaches, with determining a degree of deviation between a knowledge store and a reference knowledge store, as taught by Gelfenbeyn. The application of the determination that the degree of deviation is more than a threshold and eliminating the input in response to the determination would have been motivated to remove inputs that would interfere with convergence to a useful result. This application of a known technique from the teachings of Mara into the instructions taught by Gelfenbeyn would have yielded the predictable result that is disclosed in claim 10 of the instant application.
Regarding claim 11, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 8, and additionally, Gelfenbeyn teaches:
wherein the machine-learning model is a large language model. ((Gelfenbeyn) Paragraph [0013], “FIG. 4 is an architecture diagram that shows using a surrounding architecture of an AI character model to control an output and behavior generated by a large language model (LLM), according to an example embodiment.”)
Regarding claim 12, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 8, and additionally, Gelfenbeyn teaches:
wherein the characteristic of the media asset corresponds to a character or book title. ((Gelfenbeyn) Paragraph [0026], “In one example embodiment, the platform may receive a description of a character and generate an AI character model capable of interacting with users verbally and through emotions, gestures, actions, and movements.”)
Regarding claim 13, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 8, and additionally, Gelfenbeyn teaches:
wherein the media includes one or more strings, an image, or a video segment. ((Gelfenbeyn) Paragraph [0098], “For example, defining the scope of knowledge may include selecting, for the AI character, a canonical source or corpus of information (e.g., all Star Wars® canon) as the source of truth.” All Star Wars® canon includes books, which include strings, and movies, which include video segments.)
Regarding claim 14, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 8, and additionally, Gelfenbeyn teaches:
wherein identifying the training dataset includes defining the training dataset from data associated with the media asset and data associated with a second media asset, wherein the characteristic is included in both the data associated with the media asset and data associated with the second media asset. ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, the canonical source information may include official information, generally known information, information from a predefined source, information provided by authors of virtual characters or virtual worlds, and so forth.” Different sources of information are separate media assets. Separate media assets concerning the same canon may include the same characters.)
Regarding claim 15, Gelfenbeyn teaches:
A non-transitory computer-readable medium storing instructions that when executed by one or more processors, cause the one or more processors to perform operations including: ((Gelfenbeyn) Paragraph [0111], “The memory 1104, in some example embodiments, may refer to a non-transitory computer-readable storage medium or a computer-readable storage device… the memory 1104 is used to store program instructions for execution by the processor(s)”)
receiving an identification of a media asset;
identifying, based on the media asset, a training dataset associated with the media asset […]; ((Gelfenbeyn) Paragraph [0098], “The reference knowledge store may be associated with a canonical source information related to one of the following: a known fictional character, a known fictional world, a known environment, a known person, and so forth.” A reference knowledge store is a training dataset. A canonical source information is a media asset.)
training a machine-learning model using the training dataset ((Gelfenbeyn) Paragraph [0026], “The platform may utilize a common knowledge to train the AI character model in order to interact with the users.”), the machine-learning model being configured [to] preserve an integrity of the training dataset by detecting an unauthorized deviation between a feature vector and the training dataset; ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, defining the scope of knowledge may include comparing the knowledge store to a reference knowledge store to determine differences between the knowledge store and the reference knowledge store.” The knowledge store is a feature vector compared to the reference knowledge store which is the training dataset. Determining differences is detecting an unauthorized deviation.)
receiving a request to modify the training dataset, the request including an identification of media usable to modify the training dataset, wherein the media is associated with a characteristic of the media asset; ((Gelfenbeyn) Paragraph [0077], “An identity profile 718 selected by the user may specify elements of an AI character (e.g., role, interests) which may have an influence on how the AI character pursues goals” A user selecting an identity profile is the user sending a request to add to the knowledge store, which is modifying the training dataset. An AI character is a media asset of a media usable to modify the training dataset, and the elements of the AI character are characteristics of the media asset.)
executing the machine-learning model using a test feature vector derived at least in part from the request, wherein the machine-learning model generates an indication of a degree of deviation between the training dataset and the test feature vector; ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, defining the scope of knowledge may include comparing the knowledge store to a reference knowledge store to determine differences between the knowledge store and the reference knowledge store.” The knowledge store is a test feature vector derived from the identity profile(request) compared to the reference knowledge store which is the training dataset. Determining differences is generating an indication of a degree of deviation.)
Gelfenbeyn does not teach:
… the training dataset stored in a database that is write protected;
determining that the degree of deviation is less than a threshold; and
executing a retraining iteration of the machine-learning model in response to determining that the degree of deviation is less than the threshold.
Mara teaches:
A dataset being stored in a database that is write protected; ((Mara) Paragraph [0127], “for instance, and without limitation, temporally sequential listing may be readable and modifiable publicly, may be publicly readable but writable only by entities and/or devices having access privileges established by password protection, confidence level, or any device authentication procedure or facilities described herein, or may be readable and/or writable only by entities and/or devices having such access privileges” A dataset being stored in a database where only entities having certain access privileges to write to is being stored in a write protected database; Paragraph [0087], “In some cases, authentication datum may serve as a key to protect plurality of provider models 116 from unauthorized access and maintaining security and privacy.”)
Determining that a degree of deviation is not greater than a threshold; and ((Mara) Paragraph [0142], “‘Sanitizing’ training data, as used in this disclosure, is a process whereby training examples are removed that interfere with convergence of a machine-learning model and/or process to a useful result…a value that is more than a threshold number of standard deviations away from an average, mean, or expected value, for instance, may be eliminated.” A value that is not eliminated has been determined to not be more than a threshold degree of deviation away from a compared value.)
Executing a retraining iteration of the machine-learning model in response to accepting added training data input when the degree of deviation is not greater than the threshold. ((Mara) Paragraph [0155], “any process of training, retraining, deployment, and/or instantiation of any machine-learning model and/or algorithm may be performed and/or repeated after an initial deployment and/or instantiation to correct, refine, and/or improve the machine-learning model and/or algorithm … Event-based retraining, deployment, and/or instantiation may alternatively or additionally be triggered by receipt and/or generation of one or more new training examples” Receipt of new training examples is accepting added training data input which happens when a value is not more than a threshold number of standard deviations away from a compared value;)
Mara and Gelfenbeyn are analogous art because they are in the same area of invention: communicating with users using artificial intelligence.
Thus, it would have been obvious to a person having ordinary skill in the art before the effective filing date, having the references in front of them, to have combined storing a dataset in a write-protected database, as Mara teaches, with the identifying a training dataset from a media asset, as taught by Gelfenbeyn, and to have combined determining a degree of deviation is not greater than a threshold and executing a retraining iteration as a response to events resulting from the determination, as Mara teaches, with determining a degree of deviation between a knowledge store and a reference knowledge store, as taught by Gelfenbeyn. The application of the technique of storing the identified training dataset in a write-protected database would have been motivated to protect models from unauthorized access and to maintain security and privacy. The application of the determination that the degree of deviation is not greater than a threshold would have been motivated to check for inputs that would interfere with convergence to a useful result and executing a retraining iteration as a response to events resulting from the determination would have been motivated to refine the machine-learning model. These applications of known techniques from the teachings of Mara into the instructions taught by Gelfenbeyn would have yielded the predictable results that is disclosed in claim 15 of the instant application. The Examiner notes that this motivation applies to all dependent and/or otherwise subsequently addressed claims.
Regarding claim 16, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 15, and additionally, Gelfenbeyn teaches:
wherein the training dataset includes a set of media that represents a canon of the media asset. ((Gelfenbeyn) Paragraph [0098], “The reference knowledge store may be associated with a canonical source information related to one of the following: a known fictional character, a known fictional world, a known environment, a known person, and so forth.” A reference knowledge store is a training dataset.)
Regarding claim 17, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 15, and additionally, Gelfenbeyn teaches:
wherein the operations further include:
receiving a subsequent request to modify the training dataset; ((Gelfenbeyn) Paragraph [0077], “An identity profile 718 selected by the user may specify elements of an AI character (e.g., role, interests) which may have an influence on how the AI character pursues goals” A user selecting an identity profile is the user sending a request to add to the knowledge store, which is modifying the training dataset. An AI character is a media asset of a media usable to modify the training dataset, and the elements of the AI character are characteristics of the media asset.)
executing the machine-learning model using a new test feature vector derived from the subsequent request, wherein the machine-learning model generates a new degree of deviation between the training dataset and the new test feature vector; ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, defining the scope of knowledge may include comparing the knowledge store to a reference knowledge store to determine differences between the knowledge store and the reference knowledge store.” The knowledge store is a test feature vector derived from the identity profile(request) compared to the reference knowledge store which is the training dataset. Determining differences is generating an indication of a degree of deviation.)
Gelfenbeyn does not teach:
determining the new degree of deviation between the training dataset and the subsequent request is greater than the threshold; and
preventing the training dataset from being modified by removing the subsequent request.
Mara teaches:
Determining the degree of deviation is greater than a threshold; and
Preventing the training dataset from being modified by removing the input. ((Mara) Paragraph [0142], “‘Sanitizing’ training data, as used in this disclosure, is a process whereby training examples are removed that interfere with convergence of a machine-learning model and/or process to a useful result…a value that is more than a threshold number of standard deviations away from an average, mean, or expected value, for instance, may be eliminated.”)
Mara and Gelfenbeyn are analogous art because they are in the same area of invention: communicating with users using artificial intelligence.
Thus, it would have been obvious to a person having ordinary skill in the art before the effective filing date, having the references in front of them, to have combined determining a degree of deviation is more than a threshold and eliminating the input in response to the determination, as Mara teaches, with determining a degree of deviation between a knowledge store and a reference knowledge store, as taught by Gelfenbeyn. The application of the determination that the degree of deviation is more than a threshold and eliminating the input in response to the determination would have been motivated to remove inputs that would interfere with convergence to a useful result. This application of a known technique from the teachings of Mara into the instructions taught by Gelfenbeyn would have yielded the predictable results that is disclosed in claim 17 of the instant application.
Regarding claim 18, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 15, and additionally, Gelfenbeyn teaches:
wherein the machine-learning model is a large language model. ((Gelfenbeyn) Paragraph [0013], “FIG. 4 is an architecture diagram that shows using a surrounding architecture of an AI character model to control an output and behavior generated by a large language model (LLM), according to an example embodiment.”)
Regarding claim 19, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 15, and additionally, Gelfenbeyn teaches:
wherein the characteristic of the media asset corresponds to a character or book title. ((Gelfenbeyn) Paragraph [0026], “In one example embodiment, the platform may receive a description of a character and generate an AI character model capable of interacting with users verbally and through emotions, gestures, actions, and movements.”)
Regarding claim 20, Gelfenbeyn, in view of Mara, teaches the material disclosed in claim 15, and additionally, Gelfenbeyn teaches:
wherein identifying the training dataset includes defining the training dataset from data associated with the media asset and data associated with a second media asset, wherein the characteristic is included in both the data associated with the media asset and data associated with the second media asset. ((Gelfenbeyn) Paragraph [0098], “In an example embodiment, the canonical source information may include official information, generally known information, information from a predefined source, information provided by authors of virtual characters or virtual worlds, and so forth.” Different sources of information are separate media assets. Separate media assets concerning the same canon may include the same characters.)
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Patents and/or related publications are cited in the Notice of References Cited (Form PTO-892) attached to this action to further show the state of the art with respect to data protection, media asset analysis, and adding new training data. Any inquiry concerning this communication or earlier communications from the examiner should be directed to DYLAN H LAI whose telephone number is (571)272-8628. The examiner can normally be reached Monday - Friday 7:30am-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, Tamara Kyle can be reached at 5712524241. 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. H. L./
Examiner
Art Unit 2144
/TAMARA T KYLE/ Supervisory Patent Examiner, Art Unit 2144
Prosecution Insights