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 .
This action is responsive to Applicant’s Amendment filed on 7/8/2026.
Claims 1-20 are presented for examination. Claims 1, 6, 8 and 15 have been amended.
Applicant’s amendments to the claims have overcome claim objection set forth in the non-Final Office Action mailed 4/8/2026.
Examiner Notes
Examiner cites particular columns, paragraphs, figures and line numbers in the references as applied to the claims below for the convenience of the applicant. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and figures may apply as well. It is respectfully requested that, in preparing responses, the applicant fully consider the references in entirely as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the examiner.
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 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 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 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Birke et al. (US 20210406770 A1, hereafter Birke) in view of Keech et al. (US 20200175423 A1, hereafter Keech), Noskov (US 20240419500 A1) and Werme et al. (US 20030167270 A1, hereafter Werme).
Regarding to claim 1, Birke discloses: A system, comprising: a computing device comprising a processor and a memory; and machine-readable instructions stored in the memory that, when executed by the processor, cause the computing device to (see [0001] and [0011]; “in particular, to machine learning models used in fog networks which are implemented on an automation systems” and “A method for adjusting machine learning (ML) models in a system comprising a plurality of devices is suggested”. In order to allow the method performed in the system having plurality of devices discussed at [0011] as a automation systems discussed at [0001], it is understood that a computer device from the system discussed at [0011] to use a processor of the computer device to execute program code instructions stored at the memory of the computer device to perform the method discussed at [0011]) at least:
initiate a run-time execution of an application that includes a machine learning model (see [0011]; “providing one or more machine learning (ML) tasks” and “deploying the selected ML model to the one or more devices;—execute the selected ML model on the one or more devices”. In order to execute the selected ML associated with the provided ML tasks, it is required to initiate an run-time execution for the provided ML tasks);
determine a plurality of eligible application templates based at least in part on a machine learning model artifact for the application (see [0011]; “providing one or more machine learning (ML) tasks;—providing a repository of ML models for the one or more tasks, wherein a plurality of the ML models of a single task solve the same task with different computational resources requirements and different quality metrics”. According to the provided ML tasks, determine a plulriaty of ML models that are targeted or designed to be implemented for executing the provided ML task/application, i.e., claimed plurality of eligible application templates);
select an application template among the plurality of eligible applications templates based at least in part on run time environment data associated with an execution of a plurality of existing applications (see [0011]; “selecting one or more devices of the plurality of devices of the system to execute a task, wherein the selected one or more devices have available computational resource capacities;—selecting, from the repository of ML models of the task to be executed, one of the ML models, wherein the computational resources requirements of the selected ML model do not exceed the available computational resource capacities of the selected one or more devices”. Also see [0038]; “devices can have a primary function which uses an amount of computational resources of the device. The available computational resource capacities of the device are spare computational resource capacities which can be used by the execution of the task”), the application template comprising different computational resources and different quality metrics for executing the machine leaning model (see [0011]; “wherein a plurality of the ML models of a single task solve the same task with different computational resources requirements and different quality metrics”) and the runtime environment data comprising at least statistics data related to hardware performance metrics (see [0035]; “Based on models of the application and the fog network onto which the application should be deployed, an allocation algorithm (allocator) computes a mapping of application parts (foglets) to fog nodes. The allocation algorithm thereby can have multiple optimization objectives, for example, it shall minimize the needed network bandwidth, minimize latencies, satisfy network bandwidth constraints and constraints on latencies of data flows, and fulfill specific requirements if indicated in the application model”. The network bandwidth and latencies from [0035] are reasonable to be considered as claimed statistics data related to hardware performance metrics);
execute the application in a run-time environment specified by the application template (see [0011]; “deploying the selected ML model to the one or more devices;—execute the selected ML model on the one or more devices”).
Birke does not disclose:
the determination of plurality of eligible application templates is determining a plurality of eligible application templates based at least in part on using, as search criteria, data attributes associated with a machine learning model artifact for the application;
selecting the application template is further based on an application priority for the application; the application template comprising a computing hardware interface and a software framework.
However, Keech discloses: determining a plurality of eligible application templates based at least in part on using, as search criteria, data attributes associated with a machine learning model artifact for the application (see [0035]-[0044]; “The model subset may also be determined, at least in part, based on the datatype of the client data … The datatype of the client data includes a collection of code context corresponding to the one or more codebases. The code context may include (but are not limited to) programming language, application type”, “the client specific data 261 may indicate that the type of application being worked on by the client 270 is a mobile device application. Based on this information, the service system 210 determines that only the models that were built from mobile device project codebases are applicable to the client 270”)
It would have been obvious to one with ordinary skill, in the art before the effective filing date of the claim invention, to modify the step of providing repository of ML models for solving same task with different computational resources requirements and different quality metrics from Birke by including step of selecting a subset of ML models for certain application based on application type for the machine learning workload from Keech, since it would provide a specific mechanism to filter out inapplicable ML models to ensure providing only applicable ML models (see [0035]-[0044] from Keech; “the service system filters the collection of models 220 to include models 221 and 222 in the model subset 231 for the client 260, because the model 221 (for predicting possible APIs for Java codebases) and the model 222 (for detecting variable misuses for Java codebases) are both applicable to the datatype of the client-specific data 261 (a Java codebase). However, the model 223 (for predicting possible APIs for C # codebases) would not be applicable to the datatype of the client-specific data 261 (a Java codebase). Thus, the model 223 is not included in the model subset 231 for the client 260”).
In addition, Noskov discloses: selecting a run-time environment based on at least in part on application priority for the application and runtime environment data associated with an execution of a plurality of existing applications (see [0058]; “scheduler 116 may determine that none of node(s) 120A-120N, node(s) 122A-122N and/or cluster(s) 110A-110N have sufficient resources to satisfy deployment requirements for deploying an instance of the second service. Scheduler 116 may then evict an instance 206A-206N of a service that has a lower priority from node(s) 120A-120N, node(s) 122A-122N and/or cluster(s) 110A-110N to reallocate resources for the deployment of the first instance 208A-208N”)
It would have been obvious to one with ordinary skill, in the art before the effective filing date of the claim invention, to modify the step of selecting a ML model has computational resources requirements that do not exceed the available computational resource capacities of the selected one or more devices from the combination of Birke and Keech by including deallocating occupied resources used by an application having lower priority than the application to be deployed under a condition of none of existing nodes has sufficient available resources to meet the resource requirements of the application to be deployed from Noskov, since it would provide a mechanism to avoid a situation of none of the candidate nodes contain enough available resources to deploy the requested task while at least one candidate node is running a lower priority task (see [0058] from Noskov).
In addition, Werme discloses: the application template comprising a computing hardware interface and a software framework (see [0135], [0156]; “System Specification Files, generally denoted FG 32, which are based on this specification language, are created by the user and provide a model of the software and hardware components of the distributed computing environment which is used by the Resource Management Architecture”, “At the host level, the operating system and version, the hardware architecture, the host's network interface name”).
It would have been obvious to one with ordinary skill, in the art before the effective filing date of the claim invention, to modify the ML models specify different computational resources requirements and different quality metrics from the combination of Birke, Keech and Noskov by including specification files describe a model of software and hardware components of a distributed computing environment from Werme, and thus the combination of Birke, Keech, Noskov and Werme would disclose the missing limitations from Birke, since it would provide more specific environment data for executing tasks to group hardware and software components into systems and subsystems in order to create a hierarchy of components (see [0135]-[0146] from Werme; “the applications that can run within the distributed environment … Hardware … Hardware Configuration … Network Configuration …. Resource Requirement … QoS Requirement … the System Specification Language allows for grouping hardware and software components into systems and subsystems in order to create a hierarchy of components”).
Regarding to Claim 2, the rejection of Claim 1 is incorporated and further the combination of Birke, Keech, Noskov and Werme discloses: determine an availability of a plurality of hardware platforms based at least in part on the run time environment data for the plurality of hardware platforms (see [0011] and [0054]-[0055]; “selecting one or more devices of the plurality of devices of the system to execute a task, wherein the selected one or more devices have available computational resource capacities”).
Regarding to Claim 3, the rejection of Claim 2 is incorporated and further the combination of Birke, Keech, Noskov and Werme discloses: generate an application schedule for an execution of the application based at least in part on the availability of at least one hardware platform and the application priority for the application, wherein the execution of the application is further performed in the run-time environment based at least in part on the application schedule (see [0058] from Noskov; “scheduler 116 may determine that none of node(s) 120A-120N, node(s) 122A-122N and/or cluster(s) 110A-110N have sufficient resources to satisfy deployment requirements for deploying an instance of the second service. Scheduler 116 may then evict an instance 206A-206N of a service that has a lower priority from node(s) 120A-120N, node(s) 122A-122N and/or cluster(s) 110A-110N to reallocate resources for the deployment of the first instance 208A-208N”. Generating an application schedule that indicates that the application/service to be deployed and executed after evicting a lower priority application/service running on a node).
Regarding to Claim 4, the rejection of Claim 1 is incorporated and further the combination of Birke, Keech, Noskov and Werme discloses: generate an application schedule for an execution the application based at least in part on the application priority for the application (see [0058] from Noskov; “scheduler 116 may determine that none of node(s) 120A-120N, node(s) 122A-122N and/or cluster(s) 110A-110N have sufficient resources to satisfy deployment requirements for deploying an instance of the second service. Scheduler 116 may then evict an instance 206A-206N of a service that has a lower priority from node(s) 120A-120N, node(s) 122A-122N and/or cluster(s) 110A-110N to reallocate resources for the deployment of the first instance 208A-208N”. Generating an application schedule that indicates that the application/service to be deployed based on the application/service has a higher priority level).
Regarding to Claim 5, the rejection of Claim 4 is incorporated and further the combination of Birke, Keech, Noskov and Werme discloses: wherein the application schedule comprises an instruction to terminate the execution of a respective application on a respective hardware platform based at least in part on the application priority being higher than a respective priority of the respective application (see [0058] from Noskov; “scheduler 116 may determine that none of node(s) 120A-120N, node(s) 122A-122N and/or cluster(s) 110A-110N have sufficient resources to satisfy deployment requirements for deploying an instance of the second service. Scheduler 116 may then evict an instance 206A-206N of a service that has a lower priority from node(s) 120A-120N, node(s) 122A-122N and/or cluster(s) 110A-110N to reallocate resources for the deployment of the first instance 208A-208N”. An application schedule that indicates that the application or service to be deployed and executed after evicting or terminating a lower priority application or service running on a node).
Regarding to Claim 6, the rejection of Claim 1 is incorporated and further the combination of Birke, Keech, Noskov and Werme discloses: wherein the computing hardware interface is used by the software framework to execute at least one a portion of the application on one of a plurality of hardware platforms (see [0155]-[0156] from Werme; “network resources that the application will use at run-time” and “the host's network interface name”. Note: it is understood that the host network interface is a type of network resource that can be used by software application running on the host at run-time).
Regarding to Claim 7, the rejection of Claim 1 is incorporated and further the combination of Birke, Keech, Noskov and Werme discloses: wherein the software framework comprises a modeling framework and a distributed computing execution framework (see [0045], [0051] from Birke; “The plurality of the ML models of a single task may differ in an underlying ML algorithm” and “the task is a classification task and the plurality of the ML models include at least one or more ML models based on the random forest algorithm and/or deep neural network algorithm”. The ML task from Birke can be a classification task based on different algorithms, and thus the software framework described by the ML models or specification files must include a software modeling framework. Also see [0052] from Werme; “Specification files, based on this specification language, are created by the user and provide the model of the software and hardware components of the distributed computing environment which is used by other Resource Management functions”. The whole system described by the specification files at the combination system is a distributed computing environment having software components, and thus the software framework must include a distributed computing execution framework).
Regarding to Claim 8, Claim 8 is a method claim corresponds to system Claim 1 and is rejected for the same reason set forth in the rejection of Claim 1 above.
Regarding to Claim 9, Claim 9 is a method claim corresponds to system Claim 2 and is rejected for the same reason set forth in the rejection of Claim 2 above.
Regarding to Claim 10, Claim 10 is a method claim corresponds to system Claim 3 and is rejected for the same reason set forth in the rejection of Claim 3 above.
Regarding to Claim 11, Claim 11 is a method claim corresponds to system Claim 4 and is rejected for the same reason set forth in the rejection of Claim 4 above.
Regarding to Claim 12, Claim 12 is a method claim corresponds to system Claim 5 and is rejected for the same reason set forth in the rejection of Claim 5 above.
Regarding to Claim 13, Claim 13 is a method claim corresponds to system Claim 6 and is rejected for the same reason set forth in the rejection of Claim 6 above.
Regarding to Claim 14, Claim 14 is a method claim corresponds to system Claim 7 and is rejected for the same reason set forth in the rejection of Claim 7 above.
Regarding to Claim 15, Claim 15 is a product claim corresponds to system Claim 1 and is rejected for the same reason set forth in the rejection of Claim 1 above.
Regarding to Claim 16, Claim 16 is a product claim corresponds to system Claim 2 and is rejected for the same reason set forth in the rejection of Claim 2 above.
Regarding to Claim 17, Claim 17 is a product claim corresponds to system Claim 3 and is rejected for the same reason set forth in the rejection of Claim 3 above.
Regarding to Claim 18, Claim 18 is a product claim corresponds to system Claim 4 and is rejected for the same reason set forth in the rejection of Claim 4 above.
Regarding to Claim 19, further Claim 19 is a product claim corresponds to system Claim 5 and is rejected for the same reason set forth in the rejection of Claim 5 above.
Regarding to Claim 20, Claim 20 is a product claim corresponds to system Claim 7 and is rejected for the same reason set forth in the rejection of Claim 7 above.
Response to Arguments
Applicant’s arguments, filed 7/8/2026, with respect to rejections of claims 1-20 under 35 U.S.C. 103 have been full considered. New grounds of rejections are made based on the amended limitations from the independent claims.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Lawson et al. (US 20130212129 A1) discloses: selecting or searching a subset of object templates based on application type and then selecting one object template from the selected subset (see [0065] and [0068]).
Freese et al. (US 11669921 B2) discloses: the type of service is used to select one or more machine learning models that take the user's location and the user preferences as input to predict service providers (see claim 1).
Higgins et al. (US 20210065053 A1) discloses: based on obtaining information identifying a set of types of machine learning models and the information identifying a task type (see [0034]).
Carroll et al. (US 20230031691 A1) discloses: select a subset of the machine learning models based on the determined performance metrics. Each selected machine learning model may correspond to a different time period and/or training dataset (see [0004]).
Hamlin et al. (US 20240111610 A1) discloses: enable the selection of appropriate versions of an AI model, each version having a different level of computational complexity, for execution on corresponding devices within heterogeneous computing platform 300 based, at least in part, , upon context/telemetry data indicative of IHS resources and runtime metrics (see [0207])
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 extension fee 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 date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ZHI CHEN whose telephone number is (571)272-0805. The examiner can normally be reached on M-F from 9:30AM to 5:30PM.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, April Y Blair can be reached on 571-270-1014. 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 Patent Center and the Private Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from Patent Center or Private PAIR. Status information for unpublished applications is available through Patent Center and Private PAIR to authorized users only. Should you have questions about access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
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) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form.
/Zhi Chen/
Patent Examiner, AU2196
/APRIL Y BLAIR/Supervisory Patent Examiner, Art Unit 2196
Prosecution Insights