DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
Applicant's arguments filed 04/17/2026 have been fully considered but they are not persuasive.
Regarding the 103 rejections, applicant's arguments filed with respect to the prior art rejections have been fully considered but they are moot. Applicant has amended the claims to recite new combinations of limitations. Applicant's arguments are directed at the amendment. Please see below for new grounds of rejection, necessitated by Amendment.
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 15-20 are rejected under 35 U.S.C. 101 because the claims are not directed to one of the four categories of patent eligible subject matter.
Regarding claim 15, in step 1 of the 101 analysis set forth in MPEP 2106, the claim recites A tangible computer-readable storage media encoding computer-executable instructions for executing a computer process. A tangible computer-readable storage media is interpreted as reciting transitory signals. Therefore, the claimed invention in claim 15 is directed to non-statutory subject matter. The claim does not fall within at least one of the four categories of patent eligible subject matter because a tangible computer-readable storage media is interpreted as reciting “signals per se” which is not one of the four statutory categories (MPEP 2106.03). Applicant is encouraged to amend the claim into one of the four statutory categories.
Regarding claims 16-20, the claims are rejected for at least their dependence on claim 15.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2, 8-10, 14-15, 17, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Yellin, et al., US Pre-Grant Publication US20200285891A1 (“Yellin”) in view of Dwiveldi, et al., US Patent Publication US10754638B1 (“Dwiveldi”) and further in view of Schmidt, et al., US Pre-Grant Publication US20220156642A1 (“Schmidt”).
Regarding claim 1, Yellin discloses:
(Currently Amended) A method comprising: receiving, from a web-based service platform, a set of inputs associated with detection of a trigger event…; (Yellin, ⁋40-41, “Mobile Platform 330 is a middleware or cloud service [A method comprising: receiving, from a web-based service platform,] that performs backend operations for edge devices 310. This platform also checks the version of ML models installed on the edge devices 310 and collects various data points used in computing data sets for training the ML models. The mobile platform 330 can comprise several components. Model Training Manager 332 is a software component that continuously collects training data sets and triggers the AI System 320 to train and build new models [a set of inputs associated with detection of a trigger event…;].”).
identifying, from a configuration table, a machine learning (ML) model having execution trigger criteria…wherein the ML model is being one of multiple selectable ML models identified in the configuration table; (Yellin, ⁋42, “Model Policies 350 comprise a database table that holds information for each ML model cross-matched with known edge device types, device properties, and user characteristics [identifying, from a configuration table, a machine learning (ML) model having execution trigger criteria…wherein the ML model is being one of multiple selectable ML models identified in the configuration table;].”).
retrieving, from the configuration table, a data contract for the ML model, the data contract indicating an expected format of inputs to the ML model and an expected format of outputs generated by the ML model; (Yellin, ⁋42-43, “Model Policies 350 comprise a database table that holds information for each ML model cross-matched with known edge device types, device properties, and user characteristics. A device profile or Edge Type is a combination of both device and user characteristics, which can be expressed as a tuple <Dev Class, User Class>, where Dev Class indicates the class of device, including device attributes, and User Class indicates the type of user using the device. The Models Policies database 350 indicates for which Edge Type an ML model is known to perform well. More precisely, it associates a score for how well a model performs for an Edge Type [retrieving, from the configuration table, a data contract for the ML model, the data contract indicating an expected format of inputs to the ML model and an expected format of outputs generated by the ML model;].”).
constructing a call, based on the data contract, to the ML model, the call including input parameter values associated with the trigger event; (Yellin, ⁋47, “The software development kit (SDK) running on the edge device 310 periodically collects and sends device and user characteristics to update the device profile (Edge Type) (step 408) [constructing a call, based on the data contract,].”, and Yellin, ⁋51, “In process 600 the SDK of edge device 310 periodically uploads ML model performance to the Model Feedback component 360 on the server (step 602). The Model Orchestrator 334 periodically collects all new feedback on mode performance for all devices (step 604). The Model Orchestrator 334 performs AB testing on the feedback data to update scores (policies) in the Model Policies database 350 (step 606) [to the ML model, the call including input parameter values associated with the trigger event;].”).
While Yellin teaches a web-based platform that selects ML models from a model store, Yellin does not explicitly teach:
the trigger event being associated with one or more tasks to be performed based on the set of inputs
associated with performing the one or more tasks, the execution trigger criteria being satisfied by the set of inputs,
and translating, based on the data contract, an output received from the ML model to a unified object consumable by web-based service platform;
and providing the unified object to the web-based service platform.
Dwiveldi teaches:
and translating, based on the data contract, an output received from the ML model to a unified object consumable by web-based service platform; (Dwiveldi, col. 22 lines 12-22, “At block 512, an indexer may optionally apply one or more transformations to data included in the events created 15 at block 506. For example, such transformations can include removing a portion of an event (e.g., a portion used to define event boundaries, extraneous characters from the event, other extraneous text, etc.), masking a portion of an event (e.g., masking a credit card number), removing redundant portions of an event, etc. The transformations applied to 20 events may, for example, be specified in one or more configuration files and referenced by one or more source type definitions [and translating, based on the data contract, an output received from the ML model to a unified object consumable by web-based service platform;].”).
and providing the unified object to the web-based service platform. (Dwiveldi, col. 24 lines 19-21, “Each indexer 206 may be responsible for storing and searching a subset of events contained in a corresponding data store 208 [and providing the unified object to the web-based service platform.].”).
Yellin and Dwiveldi are both in the same field of endeavor (i.e. data processing). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin and Dwiveldi to teach the above limitation(s). The motivation for doing so is that using a specific storage structure improves the efficiency of storing large amounts of data for later retrieval (cf. Dwiveldi, col. 1 lines 38-46).
While Yellin in view of Dwiveldi teaches a web-based platform that selects ML models from a model store using a specific storage structure, the combination does not explicitly teach:
the trigger event being associated with one or more tasks to be performed based on the set of inputs
associated with performing the one or more tasks, the execution trigger criteria being satisfied by the set of inputs,
Schmidt teaches:
the trigger event being associated with one or more tasks to be performed based on the set of inputs (Schmidt, ⁋6, “The method includes generating and training, by an automated machine learning (autoML) engine, a model set including a number of different models for the ML problem. Each of the different models of the model set is specialized for a particular situation. The method further includes monitoring, by a monitoring and decision module, input data of the ML problem and selecting one or more models of the model set as active models to be applied by the resource-constrained device [the trigger event being associated with one or more tasks to be performed based on the set of inputs].”).
associated with performing the one or more tasks, the execution trigger criteria being satisfied by the set of inputs, (Schmidt, ⁋6, “The method includes generating and training, by an automated machine learning (autoML) engine, a model set including a number of different models for the ML problem. Each of the different models of the model set is specialized for a particular situation. The method further includes monitoring, by a monitoring and decision module, input data of the ML problem and selecting one or more models of the model set as active models to be applied by the resource-constrained device [associated with performing the one or more tasks, the execution trigger criteria being satisfied by the set of inputs,].”).
Yellin, in view of Dwiveldi, and Schmidt are both in the same field of endeavor (i.e. model management). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin, in view of Dwiveldi, and Schmidt to teach the above limitation(s). The motivation for doing so is that triggering the use of specific models for specific situations improves overall model accuracy (cf. Schmidt, see ⁋19).
Regarding claim 2, Yellin in view of Dwiveldi and Schmidt teaches the method of claim 1. Yellin teaches the concept of multiple selectable ML models as seen in claim 1. Dwiveldi further teaches wherein the unified object is of a consistent form regardless of which of the multiple selectable ML models generate the output. (Dwiveldi, col. 25 lines 1-3 and see Figure 5B, “FIG. 5B is a block diagram of an example data store 501 that includes a directory for each index (or partition) that contains a portion of data managed by an indexer; the information stored by the indexer is in a uniform format (i.e. wherein the unified object is of a consistent form regardless of which of the multiple selectable ML models generate the output.).”).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the present application to combine the teachings of Dwiveldi with the teachings of Yellin and Schmidt for the same reasons disclosed in claim 1.
Regarding claim 8, Yellin in view of Dwiveldi and Schmidt teaches the method of claim 1. Dwiveldi teaches the unified object as seen in claim 1. Yellin further teaches further comprising: retrieving, from the configuration table, mapping information for mapping the outputs generated by the ML model to the unified object, wherein translating the output received from the ML model depends upon the mapping information. (Yellin, ⁋51, “FIG. 6 illustrates a process flow for updating model distribution policies in accordance with an illustrative embodiment. In process 600 the SDK of edge device 310 periodically uploads ML model performance to the Model Feedback component 360 on the server (step 602). The Model Orchestrator 334 periodically collects all new feedback on mode performance for all devices (step 604). The Model Orchestrator 334 performs AB testing on the feedback data to update scores (policies) in the Model Policies database 350 (step 606) [further comprising: retrieving, from the configuration table, mapping information for mapping the outputs generated by the ML model to the unified object, wherein translating the output received from the ML model depends upon the mapping information.].”).
Regarding claim 9, the claim is similar to claim 1 and rejected under the same rationales. Yellin further discloses the additional limitations A system comprising: at least one processor; memory in electronic communication with the at least one processor; and instructions stored in the memory, the instructions being executable by the at least one processor to: (Yellin, ⁋6, “Another illustrative embodiment provides a system for distributing machine learning models to electronic devices. The system comprises a bus system, a storage device connected to the bus system, wherein the storage device stores program instructions, and a number of processors connected to the bus system, wherein the number of processors execute the program instructions to: [A system comprising: at least one processor; memory in electronic communication with the at least one processor; and instructions stored in the memory, the instructions being executable by the at least one processor to:]”).
Regarding claim 10, the claim is similar to claim 2 and is rejected under the same rationales.
Regarding claim 14, the claim is similar to claim 8 and is rejected under the same rationales.
Regarding claim 15, the claim is similar to claim 1 and rejected under the same rationales. Yellin further discloses the additional limitations A tangible computer-readable storage media encoding computer- executable instructions for executing a computer process, the computer process comprising: (Yellin, claim 15, “the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a number of processors to cause the computer to perform the steps of: [A tangible computer-readable storage media encoding computer- executable instructions for executing a computer process, the computer process comprising:]”).
Dwiveldi further teaches the additional limitation and storing the unified object in a repository accessible to the web-based service platform. (Dwiveldi, col. 24 lines 19-21, “Each indexer 206 may be responsible for storing and searching a subset of events contained in a corresponding data store 208 [and storing the unified object in a repository accessible to the web-based service platform.].”).
Yellin, in view of Schmidt, and Dwiveldi are both in the same field of endeavor (i.e. data processing). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin, in view of Schmidt, and Dwiveldi to teach the above limitation(s). The motivation for doing so is that using a specific storage structure improves the efficiency of storing large amounts of data for later retrieval (cf. Dwiveldi, col. 1 lines 38-46).
Regarding claim 17, the claim is similar to claim 8 and is rejected under the same rationales.
Regarding claim 20, Yellin in view of Dwiveldi and Schmidt teaches the tangible computer-readable storage media of claim 15. Yellin further teaches wherein the computer process further comprises: wherein the web-based service platform defines and detects the trigger event. (Yellin, ⁋40-41, “Mobile Platform 330 is a middleware or cloud service [wherein the computer process further comprises: wherein the web-based service platform] that performs backend operations for edge devices 310. This platform also checks the version of ML models installed on the edge devices 310 and collects various data points used in computing data sets for training the ML models. The mobile platform 330 can comprise several components. Model Training Manager 332 is a software component that continuously collects training data sets and triggers the AI System 320 to train and build new models [defines and detects the trigger event.].”).
Claims 3, 11, and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Yellin, et al., US Pre-Grant Publication US20200285891A1 (“Yellin”) in view of Dwiveldi, et al., US Patent Publication US10754638B1 (“Dwiveldi”) and further in view of Schmidt, et al., US Pre-Grant Publication US20220156642A1 (“Schmidt”) and Leopold, et al., US Pre-Grant Publication US20140129493A1 (“Leopold”).
Regarding claim 3, Yellin in view of Dwiveldi and Schmidt teaches the method of claim 1. While the combination teaches a web-based platform that selects ML models from a model store using a configuration table, the combination does not explicitly teach further comprising: determining, based on the configuration table, an authentication token for communicating with the ML model and a URL for accessing the ML model; and providing the authentication token to the URL.
Leopold teaches further comprising: determining, based on the configuration table, an authentication token for communicating with the ML model and a URL for accessing the ML model; and providing the authentication token to the URL. (Leopold, ⁋77, “Alternatively, in order to request a session key that does not expire, the user may be directed by the third-party application to a specific uniform or universal resource locator (“URL”) where the user may be able to generate a numerical authentication token (or “auth token”). The user may be required to provide the auth token to the third-party application. In both cases, subsequent calls to the web-based social network may return a session key that will not expire. According to a further system and method, the user may be allowed to revoke an extended or infinite session [further comprising: determining, based on the configuration table, an authentication token for communicating with the ML model and a URL for accessing the ML model; and providing the authentication token to the URL.].”).
Yellin, in view of Dwiveldi and Schmidt, and Leopold are both in the same field of endeavor (i.e. web-based data processing). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin, in view of Dwiveldi and Schmidt, and Leopold to teach the above limitation(s). The motivation for doing so is that by using authentication token with a URL ensures that the connection between a client and a server is extended (cf. Leopold, see ⁋77).
Regarding claim 11, the claim is similar to claim 3 and is rejected under the same rationales.
Regarding claim 16, the claim is similar to claim 3 and is rejected under the same rationales.
Claims 4, 12, and 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over Yellin, et al., US Pre-Grant Publication US20200285891A1 (“Yellin”) in view of Dwiveldi, et al., US Patent Publication US10754638B1 (“Dwiveldi”) and further in view of Schmidt, et al., US Pre-Grant Publication US20220156642A1 (“Schmidt”) and Chai, et al., US Pre-Grant Publication US20220091837A1 (“Chai”).
Regarding claim 4, Yellin in view of Dwiveldi and Schmidt teaches the method of claim 1. While the combination teaches a web-based platform that selects ML models from a model based on a trigger condition, the combination does not explicitly teach wherein the trigger event is detected in response to a user interaction with a user interface of the web-based service platform.
Chai teaches wherein the trigger event is detected in response to a user interaction with a user interface of the web-based service platform. (Chai, ⁋35, “According to one aspect of the present disclosure, the application development platform can enable a developer to generate custom models for their application. In one example, the developer can upload or otherwise provide access to training data and can then use the application development platform to create and train a machine-learned model for use in conjunction with their application [wherein the trigger event is detected in response to a user interaction with a user interface of the web-based service platform.].”).
Yellin, in view of Dwiveldi and Schmidt, and Chai are both in the same field of endeavor (i.e. model management). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin, in view of Dwiveldi and Schmidt, and Chai to teach the above limitation(s). The motivation for doing so is that the limitations provide comprehensive services for a model management platform (cf. Chai, see ⁋2).
Regarding claim 12, the claim is similar to claim 4 and is rejected under the same rationales.
Regarding claim 18, Yellin in view of Dwiveldi and Schmidt teaches the tangible computer-readable storage media of claim 15. While the combination teaches a web-based platform that selects ML models from a model based on a trigger condition and uses a unified object, the combination does not explicitly teach wherein the trigger event is detected in response to a user interaction with a user interface of the web-based service platform and the unified object is automatically rendered to the user interface.
Chai teaches wherein the trigger event is detected in response to a user interaction with a user interface of the web-based service platform and the unified object is automatically rendered to the user interface. (Chai, ⁋35, “According to one aspect of the present disclosure, the application development platform can enable a developer to generate custom models for their application. In one example, the developer can upload or otherwise provide access to training data and can then use the application development platform to create and train a machine-learned model for use in conjunction with their application [wherein the trigger event is detected in response to a user interaction with a user interface of the web-based service platform].”, and Chai, ⁋176, “FIG. 17 depicts an example user interface according to example embodiments of the present disclosure. The user interface can be referred to as a Tensorboard. Once the training starts, users can monitor a training progress on Tensorboard. The Tensorboard has several options (e.g., scalars, graphs, distributions, histograms, and projectors) to present the training progress [and the unified object is automatically rendered to the user interface.].”).
Yellin, in view of Dwiveldi and Schmidt, and Chai are both in the same field of endeavor (i.e. model management). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin, in view of Dwiveldi and Schmidt, and Chai to teach the above limitation(s). The motivation for doing so is that the limitations provide comprehensive services for a model management platform (cf. Chai, see ⁋2).
Regarding claim 19, Yellin in view of Dwiveldi and Schmidt teaches the tangible computer-readable storage media of claim 15. While the combination teaches a web-based platform that selects ML models from a model based on a trigger condition and uses a unified object, the combination does not explicitly teach wherein the computer process further comprises: in response to storing the unified object in the repository, automatically rendering the unified object to a user interface of the web-based service platform.
Chai teaches wherein the computer process further comprises: in response to storing the unified object in the repository, automatically rendering the unified object to a user interface of the web-based service platform. (Chai, ⁋35, “According to one aspect of the present disclosure, the application development platform can enable a developer to generate custom models for their application. In one example, the developer can upload or otherwise provide access to training data and can then use the application development platform to create and train a machine-learned model for use in conjunction with their application; selecting the model to be used is in response to storing a model selection (i.e. wherein the computer process further comprises: in response to storing the unified object in the repository,).”, and Chai, ⁋176, “FIG. 17 depicts an example user interface according to example embodiments of the present disclosure. The user interface can be referred to as a Tensorboard. Once the training starts, users can monitor a training progress on Tensorboard. The Tensorboard has several options (e.g., scalars, graphs, distributions, histograms, and projectors) to present the training progress [automatically rendering the unified object to a user interface of the web-based service platform.].”).
Yellin, in view of Dwiveldi and Schmidt, and Chai are both in the same field of endeavor (i.e. model management). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin, in view of Dwiveldi and Schmidt, and Chai to teach the above limitation(s). The motivation for doing so is that the limitations provide comprehensive services for a model management platform (cf. Chai, see ⁋2).
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Yellin, et al., US Pre-Grant Publication US20200285891A1 (“Yellin”) in view of Dwiveldi, et al., US Patent Publication US10754638B1 (“Dwiveldi”) and further in view of Schmidt, et al., US Pre-Grant Publication US20220156642A1 (“Schmidt”), Leopold, et al., US Pre-Grant Publication US20140129493A1 (“Leopold”), and Chai, et al., US Pre-Grant Publication US20220091837A1 (“Chai”).
Regarding claim 5, Yellin in view of Dwiveldi, Schmidt, and Leopold teaches the method of claim 3. While the combination teaches a web-based platform that selects ML models from a model store based on a configuration table and storing unified objects in repository, the combination does not explicitly teach further comprising: accessing the configuration table using a first application programming interface (API) and placing the unified object in a repository using a second API, the repository being accessible to one or more user interfaces of the web-based service platform.
Chai teaches further comprising: accessing the configuration table using a first application programming interface (API) and placing the unified object in a repository using a second API, the repository being accessible to one or more user interfaces of the web-based service platform. (Chai, ⁋78, “In one example, the developer may be enabled to specify (e.g., for a subset of devices and/or models) whether inference and/or training occurs on-device or via a cloud service. Both of these options can be supported by a single machine intelligence SDK, providing dynamically-controllable flexibility around the location of inference/training. In some implementations, different APIs or parameters can be used to support each of the two options. In some implementations, the developer can specify rules that handle when inference and/or training should occur on-device or in the cloud and the machine intelligence SDK can implement these rules [further comprising: accessing the configuration table using a first application programming interface (API)].”, and Chai, ⁋189, “FIG. 5 depicts a workflow diagram of an example computing system according to example embodiments of the present disclosure. In particular, FIG. 5 depicts an example of a basic device-side flow. In particular, the application development platform and machine intelligence SDK allow developers to use custom models on the device. The API allows the application to perform inference using a custom model that's already on the device, where the developer either manually downloads the model file from somewhere or packages it within the application itself [and placing the unified object in a repository using a second API, the repository being accessible to one or more user interfaces of the web-based service platform.].”).
Yellin, in view of Dwiveldi, Schmidt, and Leopold, and Chai are both in the same field of endeavor (i.e. model management). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin, in view of Dwiveldi, Schmidt, and Leopold, and Chai to teach the above limitation(s). The motivation for doing so is that the limitations provide comprehensive services for a model management platform (cf. Chai, see ⁋2).
Claims 6-7 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Yellin, et al., US Pre-Grant Publication US20200285891A1 (“Yellin”) in view of Dwiveldi, et al., US Patent Publication US10754638B1 (“Dwiveldi”) and further in view of Schmidt, et al., US Pre-Grant Publication US20220156642A1 (“Schmidt”) and Schott, et al., US Pre-Grant Publication US20210256427A1 (“Schott”).
Regarding claim 6, Yellin in view of Dwiveldi and Schmidt teaches the method of claim 1. While the combination teaches a web-based platform that selects ML models based on a configuration table and translates an output received from a model, the combination does not explicitly teach wherein the ML model executes asynchronously from a processing thread that constructs the call to the ML model and wherein translating the output received from the ML model is performed in response to a call initiated by the ML model.
Schott teaches wherein the ML model executes asynchronously from a processing thread that constructs the call to the ML model and wherein translating the output received from the ML model is performed in response to a call initiated by the ML model. (Schott, ⁋54, “inferences (which may include inference values) that are written to the inference file(s) retain their previous values until written again. As such, a kernel component or non-kernel component seeking inference values, can read the inference values from the inference file(s) without waiting for a new inference; that is, inference daemon 130 and trained machine learning model(s) 132 can write inferences to the inference file(s) and kernel components and/or non-kernel components can read the inferences from the inference file(s) asynchronously [wherein the ML model executes asynchronously from a processing thread that constructs the call to the ML model and wherein translating the output received from the ML model is performed in response to a call initiated by the ML model.].”).
Yellin, in view of Dwiveldi and Schmidt, and Schott are both in the same field of endeavor (i.e. machine learning). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin, in view of Dwiveldi and Schmidt, and Schott to teach the above limitation(s). The motivation for doing so is that using asynchronous calls reduces waiting time (cf. Schott, see ⁋54).
Regarding claim 7, Yellin in view of Dwiveldi and Schmidt teaches the method of claim 1. While the combination teaches a web-based platform that selects ML models based on a configuration table and translates an output received from a model, the combination does not explicitly teach wherein the ML model executes synchronously with a processing thread that constructs the call to the ML model and wherein translating the output received from the ML model is performed in response to receiving the output from the ML model.
Schott teaches wherein the ML model executes synchronously with a processing thread that constructs the call to the ML model and wherein translating the output received from the ML model is performed in response to receiving the output from the ML model. (Schott, ⁋56, “In other examples, a synchronized interface can deliver inferences from trained machine learning model(s) 132 at predetermined time intervals and/or within a predetermined amount of time from requesting an inference—such a synchronized interface can be used along with or instead of asynchronous inferences [wherein the ML model executes synchronously with a processing thread that constructs the call to the ML model and wherein translating the output received from the ML model is performed in response to receiving the output from the ML model.].”).
Yellin, in view of Dwiveldi and Schmidt, and Schott are both in the same field of endeavor (i.e. machine learning). It would have been obvious for a person having ordinary skill in the art before the effective filing date of the claimed invention to combine Yellin, in view of Dwiveldi and Schmidt, and Schott to teach the above limitation(s). The motivation for doing so is that using synchronous calls ensures that data is received in predetermined time intervals (cf. Schott, see ⁋56).
Regarding claim 13, the claim is similar to claim 6 and is rejected under the same rationales.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Magasweran, et al., US12423209B1 discloses a system that detects whether a device has a changed state or resource availability and whether a ML model cache or system memory needs to be resized to accommodate new ML models for the changed state.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NICHOLAS S WU whose telephone number is (571)270-0939. The examiner can normally be reached Monday - Friday 8:00 am - 4:00 pm EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Michelle Bechtold can be reached at 571-431-0762. 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.
/N.S.W./Examiner, Art Unit 2148 /MICHELLE T BECHTOLD/Supervisory Patent Examiner, Art Unit 2148
Prosecution Insights