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 .
DETAILED ACTION
This Office Action is sent in response to Application’s Communication received on 10/11/2023 for application number 18/483266. The Office hereby acknowledges receipt of the following and placed of record in file: Specification, Drawing, Abstract, Oath/Declaration, and Claims.
Claims (1-14), (15-19) and 20 are presented for examination.
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 03/23/2026, 11/17/2025, 07/07/2025, 04/08/2025 and 04/24/2024 were filed prior to current Office Action. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 3, 6-7, 9, 11-12, 15, 17 and 20 are rejected under AIA 35 U.S.C. 103(a) as being unpatentable over of Vogeti et al. US Patent Application Publication US 20220129785 (hereinafter Vogeti) in view of Kurniawan et al. US Patent Application Publication US 11449797 B1 (hereinafter Kurniawan)
Regarding claim 1, Vogeti teaches A computer-implemented method comprising: storing, by one or more processors, at least one machine learning model trained utilizing at least one third-party workspace; initiating, by the one or more processors, a deployed instance of a selected machine learning model, wherein the deployed instance of the selected machine learning model is operable via a first-party workspace ([0014], [0016], [0043], [0051], [0078] wherein Vogeti teaches a service provider, which may provide services to users and service provider server with ML model platform and packages that provides an interface for connecting users, selecting and deploying models and wherein ML model packages include model artifacts, wherein the service provider includes processors) receiving, by the one or more processors, at least one data artifact in response to operation of the deployed instance of the selected machine learning model via the first-party workspace ([0019], [0023], [0035] wherein Vogeti provides a convenient interface that permit users for client device view, and/or utilize ML models and other AI services for an intelligent platform of service provider server. Wherein the user interface (UI) of an ML model uploader may present one or more fields and operations for uploading and deployment of an ML model to an ML prediction engine of service provider server. This may include a data package that includes the model artifacts necessary to deploy the ML model, such as the trained layers, nodes, weights, values, and/or classifiers for the ML model).
Vogeti does not teach generating, by the one or more processors, updated evaluation data associated with the deployed instance of the selected machine learning model based on the at least one data artifact; determining, by the one or more processors, that the updated evaluation data does not satisfy at least one model maintenance threshold; and triggering, via the one or more processors, a process that terminates access to at least the deployed instance of the selected machine learning model in response to determining that the updated evaluation data does not satisfy the at least one model maintenance threshold.
However in analogous art of automated deployed governance, Kurniawan teaches generating, by the one or more processors, updated evaluation data associated with the deployed instance of the selected machine learning model based on the at least one data artifact; determining, by the one or more processors, that the updated evaluation data does not satisfy at least one model maintenance threshold (¶ 23, ¶ 27, ¶ 30, ¶ 35-36, wherein Kurniawan describes the training and evaluation coordinators of a secure machine learning automation service (SMLAS) that may be implemented at a provider network to support such functionality and that may utilize the artifacts to automatically generate a representation of a dynamically deployable software execution environment which satisfies the software dependencies and includes the training programs. Wherein describes the secure interfaces and an instance of the dynamically deployable software execution environment and trains and computes the instance at the model that is trained. Wherein the newly trained version of the model may be evaluated, wherein a client may use such secure interfaces to examine/explore the training-related artifacts and gain confidence in the quality of the model. The resources set up for the exploration of the artifacts may be automatically decommissioned or terminated by the SMLAS so as to prevent any unintentional data leaks in various embodiments) and triggering, via the one or more processors, a process that terminates access to at least the deployed instance of the selected machine learning model in response to determining that the updated evaluation data does not satisfy the at least one model maintenance threshold (¶ 12, ¶ 30, ¶ 50, ¶ wherein Kurniawan describes evaluating and a cleaning method that includes terminating decommissioning models based metrics).
It would have been obvious to a person in the ordinary skill in the art before the effective filing date of the claimed invention to combine Vogeti with Kurniawan by incorporating the method of generating, by the one or more processors, updated evaluation data associated with the deployed instance of the selected machine learning model based on the at least one data artifact; determining, by the one or more processors, that the updated evaluation data does not satisfy at least one model maintenance threshold; and triggering, via the one or more processors, a process that terminates access to at least the deployed instance of the selected machine learning model in response to determining that the updated evaluation data does not satisfy the at least one model maintenance threshold of Kurniawan into the method of receiving, by the one or more processors, at least one data artifact in response to operation of the deployed instance of the selected machine learning model via the first-party workspace of Vogeti for the purpose of correctness of the deployment (e.g., whether the input data sources for whose records results are to be generated using the model have been configured to transmit the data records to the right platforms, whether the model output is being routed to the appropriate destinations, etc.) may be verified in stage (Kurniawan: ¶ 36).
Regarding claim 3, Vogeti as modified by Kurniawan teach initiating, by the one or more processors, a deployment workspace that provides access to the deployed instance of the selected machine learning model, wherein the deployment workspace is configured to enable use or further training of the selected machine learning model; and configuring the deployment workspace to be accessible via the first-party workspace (FIG. 1, FIGS. 2-3B, [0050], [0071] wherein Vogeti describes an application that contains software programs, executable by a processor, including a graphical user interface (GUI), configured to provide an interface to the user when accessing service provider server, where the user or other users may interact with the GUI to more easily view and communicate information. Wherein the application includes additional connection and/or communication applications, which may be utilized to communicate information to over network. Wherein is users can interact with a ModelConfigServ of web server in order to provide model deployment services. Wherein the ModelConfigServ may correspond to a configuration service, resource, and/or object that allows authorized users to upload ML models, validate those ML models, and thereafter deploy those models for ML predictions using the ML models, for example, as illustrated in FIGS. 2-3B. ModelConfigServ may then store data for the ML models, such as metadata and/or model artifacts, to a model repository, that may include one or more databases or data stores for this metadata and artifacts. Thereafter, web server may provide one or more UIs and/or services for utilising these ML models.
Regarding claim 6, Vogeti as modified by Kurniawan teach wherein the deployed instance of the selected machine learning model is initiated utilizing at least one third-party workspace accessible via the first-party workspace, and wherein receiving the at least one data artifact in response to operation of the deployed instance comprises: receiving, by the one or more processors, the at least one data artifact via at least one workspace data hook that receives the at least one data artifact via the at least one third-party workspace upon updated publication of the deployed instance of the selected machine learning model (FIG. 2, [0076-0057] wherein Vogeti provides an application with functionalities to create and/or delete service nodes for model deployments and fetch deployment information. Wherein the application includes creating a service node that allows for deployment of an ML model, viewing of validation/verification status, and deployment status once the ML model is validated. The application may correspond to a central store for live information related to the ML prediction service, engine, and platform, which includes information about current status of model deployments for each instance of the ML model engine provided to users, such as user 202 in FIG. 2. The application describes GPE 210 that may fetch models to deploy and metadata for those models, at step 5. GPE 210 may correspond to a generic prediction engine, such as an ML prediction engine for multiple different ML models, including those previously deployed and currently uploaded and verified for deployment. Thus, GPE 210 may interact with zookeeper 212 in order to provide ML model deployment statuses and determine any deployment requests, as well as to assist in providing instances of the ML model deployer and ML prediction engine, and corresponding data, to users including user. Wherein ML model artifacts in model artifact repository 208 may be accessed and used when the ML model is validated, as well as during later predictive services).
Regarding claim 7, Vogeti as modified by Kurniawan teach wherein the deployed instance of the selected machine learning model is initiated utilizing at least one third-party workspace accessible via the first-party workspace, and wherein receiving the at least one data artifact in response to operation of the deployed instance comprises: receiving, by the one or more processors, the at least one data artifact via at least one workspace data hook that retrieves the at least one data artifact via the at least one third-party workspace in real-time in response to the operation of the deployed instance of the selected machine learning model ([0023], [0069], [0081] wherein Vogeti teaches retrieving the model artifacts from model artifact repository, this may be done in real-time. Wherein Vogeti describes removing ML models from predictive services, such as if a model is detected as behaving incorrectly (e.g., not providing expected predictions) or has errors).
Regarding claim 9, Vogeti as modified by Kurniawan teach determining, by the one or more processors, that at least one metric value of the updated evaluation data does not satisfy a metric minimum threshold defined by a minimum evaluation threshold of the at least one model maintenance threshold (¶ 12, ¶ 30, ¶ 50, ¶ wherein Kurniawan describes evaluating and a cleaning method that includes terminating decommissioning models-based metrics).
Regarding claim 11, Vogeti as modified by Kurniawan teach wherein triggering the process that terminates access to at least the deployed instance of the selected machine learning model comprises: configuring, by the one or more processors, the first-party workspace to make at least the deployed instance of the selected machine learning model inaccessible to a particular user profile associated with the first-party workspace ([0014], [0018], [0082] wherein Vogeti allows users to access machine learning models based on their registration and profiles).
Regarding claim 12, Vogeti as modified by Kurniawan teach wherein triggering the process that terminates access to at least the deployed instance of the selected machine learning model comprises: configuring, by the one or more processors, at least the first workspace to make a plurality of deployed instances associated with the selected machine learning model inaccessible, wherein the plurality of deployed instances are associated with a plurality of user profiles ([0014], [0018], [0082] wherein Vogeti provides a machine learning models platforms that is configured to users to access machine learning models based on their registration and profiles).
Regarding claim 15, Vogeti teaches a system comprising at least one memory and one or more processors communicatively coupled to the at least one memory, the one or more processors configured ([0032]). The claim is similar in scope to claim 1 therefore the claims are rejected under similar rationale.
Regarding claim 17, the claim is similar in scope to claim 3 therefore the claims are rejected under similar rationale.
Regarding claim 20, Vogeti teaches at least one non-transitory computer-readable storage medium having instructions that, when executed by at least one processor, cause the at least one processor to ([0087-0088]). The claim is similar in scope to claim 20 therefore the claims are rejected under similar rationale.
Claims 2, 16 and 18 are rejected under AIA 35 U.S.C. 103(a) as being unpatentable over of Vogeti et al. US Patent Application Publication US 20220129785 (hereinafter Vogeti) in view of Kurniawan et al. US Patent Application Publication US 11449797 B1 (hereinafter Kurniawan) and further in view of Lee et al. US Patent Application Publication US 20230111775 A1 (hereinafter Lee).
Regrading claim 2, Vogeti and Kurniawan does not teach receiving, by the one or more processors, user engagement indicating the selected machine learning model from the at least one stored machine learning model in response to a search query executed that results in retrieval of the at least one machine learning model.
However in analogous art of automated deployed governance, Lee teaches receiving, by the one or more processors, user engagement indicating the selected machine learning model from the at least one stored machine learning model in response to a search query executed that results in retrieval of the at least one machine learning model (FIG. 6, Claims 3, 5, 13 and 15, [0097] wherein Lee describes querying the model registry to locate information related a machine learning model).
It would have been obvious to a person in the ordinary skill in the art before the effective filing date of the claimed invention to combine Lee with Vogeti and Kurniawan by incorporating the method of receiving, by the one or more processors, user engagement indicating the selected machine learning model from the at least one stored machine learning model in response to a search query executed that results in retrieval of the at least one machine learning model of Lee into the method of receiving, by the one or more processors, at least one data artifact in response to operation of the deployed instance of the selected machine learning model via the first-party workspace of Vogeti and Kurniawan for the purpose of receiving information related to the model tag from the querying, the information comprising respective model tags associated with each machine learning model from the set of dependent machine learning models (Lee: Claim 13 text ).
Regarding claim 16, the claim is similar in scope to claim 2 therefore the claims are rejected under similar rationale.
Regarding claim 8, Vogeti as modified by Kurniawan and Lee teach wherein the operation of the deployed instance of the selected machine learning model comprises updated training of the deployed instance of the selected machine learning model or use of the deployed instance of the selected machine learning model for a data processing task ([0019], [0035], [0065], [0069-0074] wherein Lee describes steps of updating and deploying the training models).
Claim 4 is rejected under AIA 35 U.S.C. 103(a) as being unpatentable over of Vogeti et al. US Patent Application Publication US 20220129785 (hereinafter Vogeti) in view of Kurniawan et al. US Patent Application Publication US 11449797 B1 (hereinafter Kurniawan) and further in view of Wang. US Patent Application Publication US 20220012030 A1 (hereinafter Wang).
Regarding claim 4, Vogeti and Kurniawan do not teach wherein the deployment workspace comprises a sub-workspace of the first-party workspace or at least one additional third-party workspace.
However in analogous art of automated deployed governance, Wang teaches wherein the deployment workspace comprises a sub-workspace of the first-party workspace or at least one additional third-party workspace (FIG. 2, Abstract, [0010], [0019-0020] wherein Wang describes a system and method are described for creating application-related infrastructure resources from an application deployment platform (ADP), but which can have a single audit trail and common enforcement point of policies. A workspace custom resource definition (CRD) is generated to define a workspace schema for the workspace. The workspace schema represents a collection of configurations and variables for operating the infrastructure resources. An infrastructure controller (IC) operator is provided to the ADP to extend the API for communication with an infrastructure controller (IC), which has a set of IC definitions that define the infrastructure resources for the workspace. The workspace is built with the infrastructure resources defined by a workspace custom resource, and the CRD is deployed to the ADP via the IC operator to create the workspace custom resource based on the collection of configurations and the one or more variables. Wang further describes creating and building a workspace custom resource based on the collection of configurations and the one or more variables, enabling the ADP 102 to deploy the workspace with the infrastructure resources defined by the workspace custom resource).
It would have been obvious to a person in the ordinary skill in the art before the effective filing date of the claimed invention to combine Wang with Vogeti and Kurniawan by incorporating the method of wherein the deployment workspace comprises a sub-workspace of the first-party workspace or at least one additional third-party workspace of Wang into the method of receiving, by the one or more processors, at least one data artifact in response to operation of the deployed instance of the selected machine learning model via the first-party workspace of Vogeti and Kurniawan for the purpose configuring to retrieve values from the workspace definition, create or update a workspace, create or update variables in the workspace, and update a status or state of the workplace in the ADP (Wang: [0023]).
Claims 5 and 18 are rejected under AIA 35 U.S.C. 103(a) as being unpatentable over of Vogeti et al. US Patent Application Publication US 20220129785 (hereinafter Vogeti) in view of Kurniawan et al. US Patent Application Publication US 11449797 B1 (hereinafter Kurniawan) and further in view of Ravi. US Patent Application Publication US 20230267373 A1 (hereinafter Ravi).
Regarding claim 5, Vogeti and Kurniawan do not teach wherein the deployed instance of the selected machine learning model comprises a first deployed instance operable by at least a first user profile, and wherein the computer-implemented method further comprises: initiating, by the one or more processors, a second deployed instance of the selected machine learning model, wherein the second deployed instance is operable by at least a second user profile, wherein the first deployed instance and the second deployed instance are independently operable.
However in analogous art of automated deployed governance, Ravi teaches wherein the deployment workspace comprises a sub-workspace of the first-party workspace or at least one additional third-party workspace (FIGS. 2-FIG.S 3A-3B, FIG. 4, 6A-6B, 6C-6D, FIGS. 7-8D, [0078] wherein Ravi describes under certain circumstances, multiple different users may train and personalize the machine learning models on individual data and specific applications to produce different and private versions of the AI engine tailored to their individual needs. For example, user A may train a topic detection system to narrow down and recognize posts specifically related to “baseball,” whereas user B may train the engine to surface tennis-related posts instead. All other generic users may see posts about broader topics (e.g., sports or politics in general) if the AI engine is not trained with personal information for these users. As another example, an enterprise with multiple divisions may use the privacy engine 218 to train different AI engines (i.e., different versions of a machine learning model) that cater to different cohorts of customers. The enterprise may then choose to grant/deny certain divisions access to the AI engine in a selective manner using the private sharing option. For example, a customer service chatbot can be trained and accessed by the finance division to help with banking or payment issues, whereas the same chatbot can be also trained and used by the IT division to respond differently to IT/technical issues. As a result, once trained, the AI engine is automatically and natively privacy-preserving. The specific processes of training a personalized AI engine are further described in detail in FIGS. 7-8D).
It would have been obvious to a person in the ordinary skill in the art before the effective filing date of the claimed invention to combine Ravi with Vogeti and Kurniawan by incorporating the method of wherein the deployment workspace comprises a sub-workspace of the first-party workspace or at least one additional third-party workspace of Ravi into the method of receiving, by the one or more processors, at least one data artifact in response to operation of the deployed instance of the selected machine learning model via the first-party workspace of Vogeti and Kurniawan for the purpose of training a personalized AI engine for users (Ravi: [0078]).
Regarding claim 18, the claim is similar in scope to claim 5 therefore the claims are rejected under similar rationale.
Claims 10, 13-14 and 19 are rejected under AIA 35 U.S.C. 103(a) as being unpatentable over of Vogeti et al. US Patent Application Publication US 20220129785 (hereinafter Vogeti) in view of Kurniawan et al. US Patent Application Publication US 11449797 B1 (hereinafter Kurniawan) and further in view of Cella. US Patent Application Publication US 20230196230 A1 (hereinafter Cella).
Regarding claim 10, Vogeti and Kurniawan do not teach wherein determining that the updated evaluation data does not satisfy the at least one model maintenance threshold comprises: determining, by the one or more processors, a drift metric data based on the updated evaluation data; and determining, by the one or more processors, that the drift metric data indicates an unacceptable data drift based on a drift threshold of the at least one model maintenance threshold.
However in analogous art of automated deployed governance, Cella teaches wherein determining that the updated evaluation data does not satisfy the at least one model maintenance threshold comprises: determining, by the one or more processors, a drift metric data based on the updated evaluation data; and determining, by the one or more processors, that the drift metric data indicates an unacceptable data drift based on a drift threshold of the at least one model maintenance threshold ([0454], [0545], [1233], [1735], [1776], [1895], [3888], [3934], [4075], [4642] wherein Cella describes sensor data identifiers of each sensor measurement and a timestamp of each sensor measurement or group of sensor measurements) wherein a block may store sensor measurements determined to be anomalous (e.g., outside a standard deviation of expected sensor measurements or deltas in sensor measurements that are above a threshold) and/or sensor measurements indicative of an issue or potential issue, and related metadata (e.g., sensor IDs of each sensor measurement and a timestamp of each sensor measurement or group of sensor measurements. Wherein Cella determines a maintenance triggering threshold for operational data, including sensed data, may include identifying a type of effect the data represents and then determining data values that represent acceptable operation, questionable operation, unacceptable operation, and other types of operation. In embodiments, vibration sensors deployed to detect and monitor vibration activity of industrial machine components, structural elements, and the like may facilitate determining how vibration of machine parts contributes to predictive maintenance actions).
It would have been obvious to a person in the ordinary skill in the art before the effective filing date of the claimed invention to combine Cella with Vogeti and Kurniawan by incorporating the method of wherein determining that the updated evaluation data does not satisfy the at least one model maintenance threshold comprises: determining, by the one or more processors, a drift metric data based on the updated evaluation data; and determining, by the one or more processors, that the drift metric data indicates an unacceptable data drift based on a drift threshold of the at least one model maintenance threshold of Cella into the method of receiving, by the one or more processors, at least one data artifact in response to operation of the deployed instance of the selected machine learning model via the first-party workspace of Vogeti and Kurniawan for the purpose of incorporating system for predicting maintenance events for industrial machines may apply severity/action unit values for industrial machines in a simple comparison function that compares a severity/action unit value to the severity/action unit threshold value. When the unit value is below the threshold value, an impact on a prediction of a need for maintenance may be small or negligible (Cella: [3934]).
Regarding claim 13, Vogeti as modified by Kurniawan and Cella teach causing rendering, by the one or more processors, of a notification comprising a prompt for each user profile of the plurality of user profiles to initiate a new sub-workspace configured to enable updated training of a new instance the selected machine learning model and publication of the new instance upon completion of the updated training ([0065], [1064], [1070], [1083] wherein Cella Triggers alert initiating maintenance and alignment and deploying a field technician; recommending a vibration absorption/dampening device; modifying a process to utilize backup equipment/component; modifying a process to preserve products/reactants).
Regarding claim 14, Vogeti as modified by Kurniawan and Cella teach causing rendering, by the one or more processors, of a notification comprising a prompt for a user to initiate a new sub-workspace configured to enable updated training of a new instance the selected machine learning model and publication of the new instance upon completion of the updated training ([0065], [1064], [1070], [1083] wherein Cella Triggers alert initiating maintenance and alignment and deploying a field technician; recommending a vibration absorption/dampening device; modifying a process to utilize backup equipment/component; modifying a process to preserve products/reactants).
Regarding claim 19, the claim is similar in scope to claim 10 therefore the claims are rejected under similar rationale.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HASSAN MRABI whose telephone number is (571)272-8875. The examiner can normally be reached on Monday-Friday, 7:30am-5pm. Alt, Friday, EST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Viker Lamardo can be reached on 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 an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/HASSAN MRABI/Examiner, Art Unit 2144
Prosecution Insights