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 2/12/2026 have been fully considered but they are not persuasive.
Regarding applicant’s argument for U.S.C. 103, applicant argues on page 8-9 “Although Applicant respectfully disagrees with the rejection under 35 U.S.C. § 103, for the purpose of expediting allowance of the present application, Applicant has amended claim 1. Specifically, claim 1 has been amended and recites: … Applicant respectfully submits that Horesh and Finnie, individually or in combination, fails to teach at least in part the feature of … as recited in amended claim 1.” Applicant argues how Haresh and Finnie combined failed to teach the amended limitations. However the amended limitations have not been examined and render the argument moot.
Examiner’s Remarks
The Examiner notes that the instant application was previously examined by a different examiner. As such, the Examiner proceeds with prosecution giving full faith and credit to the search and action of the previous examiner per MPEP § 704.01:
When an examiner is assigned to act on an application which has received one or more actions by some other examiner, full faith and credit should be given to the search and action of the previous examiner unless there is a clear error in the previous action or knowledge of other prior art. In general the second examiner should not take an entirely new approach to the application or attempt to reorient the point of view of the previous examiner, or make a new search in the mere hope of finding something. See MPEP § 719.05.
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 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Horesh et al. (US20220129771A1) (“Horesh”) in view of Yuan et al. (US11599813B1) (“Yuan”).
Regarding claim 1, Horesh teaches receiving a request to perform inferencing [using a specified machine learning model] (Horesh Fig. 2,
Para 0064 line 1-7, As illustrated , operations 200 begin at block 210 , where a system receives a request to perform an inference with respect to a set of input data . The system can receive the request to perform an inference with respect to a set of input data from a requesting service , for example , when input data is received through user input , from another computing system , or otherwise ingested [A method, comprising: receiving a request to perform inferencing]);
identifying, in a distributed computing environment, a plurality of resources available to perform the inferencing (Horesh para 0065 line1-8, At block 220 , the system selects one or more client devices to be used in performing the inference with respect to the set of input data . The system can select one or more devices to be used in performing the inference based on a list of devices that are enrolled with a central server as candidate devices [a plurality of resources available to perform the inferencing] that can participate in a distributed inference generation procedure within a distributed computing environment [identifying, in a distributed computing environment].);
performing the inferencing using the [specified] machine learning model with the selected combination of the available resources of the distributed computing environment;
and providing one or more results of the inferencing in response to the request (Horesh para 0067, At block 240 , the system receives an inference response from each of the selected one or more client devices . The inference response received from each of the selected one or more client devices may include , for example , an inference generated by a client device ( e.g. , a category to which the input data set belongs ) and a confidence level associated with the inference [performing the inferencing using the [specified] machine learning model with the selected combination of the available resources of the distributed computing environment].
Para 0069, At block 260 , the system outputs the aggregated inference response to a requesting service . The aggregated inference response may be output to another application executing on the system or to an application executing on a remote computing system [and providing one or more results of the inferencing in response to the request].).
Horesh does not explicitly teach
[receiving a request to perform inferencing] using a specified machine learning model
determining a set of requirements associated with the specified machine learning model, the set of requirements characterizing one or more operational parameters of execution of the specified machine learning model to perform the inferencing;
selecting, based at least on the set of requirements and a set of resource compatibility information characterizing capabilities of the plurality of resources relative to the set of requirements, a combination of [[the]] available [[of]] resources of the distributed computing environment to perform the inferencing;
[performing the inferencing using the] specified [machine learning model with the selected combination of the available resources of the distributed computing environment];
However Yuan teaches [receiving a request to perform inferencing] using a specified machine learning model (Yuan col 11 line 22-32, The interactive workflow builder may determine additional prompt(s) to solicit configuration information for customizing a workflow template for the specific needs of the user. For example, the user interface may solicit input representing a training input dataset, references for a data collection step, data quality rules for a quality monitoring step, a specific model training algorithm for a training step, model approval rules for an approval step, locations at which to store outputs of various steps, a specific version of a model to be used for inference [using a specified machine learning model], and so on);
determining a set of requirements associated with the specified machine learning model, the set of requirements characterizing one or more operational parameters of execution of the specified machine learning model to perform the inferencing (Yuan col 11 line 22-32, The interactive workflow builder may determine additional prompt(s) to solicit configuration information for customizing a workflow template for the specific needs of the user. For example, the user interface may solicit input representing a training input dataset, references for a data collection step, data quality rules for a quality monitoring step, a specific model training algorithm for a training step, model approval rules for an approval step, locations at which to store outputs of various steps, a specific version of a model to be used for inference, and so on [determining a set of requirements associated with the specified machine learning model] [the set of requirements characterizing one or more operational parameters of execution of the specified machine learning model to perform the inferencing]);
selecting, based at least on the set of requirements and a set of resource compatibility information characterizing capabilities of the plurality of resources relative to the set of requirements, a combination of [[the]] available [[of]] resources of the distributed computing environment to perform the inferencing (Yuan Col 10 line 1-4, In one embodiment, components of the system 100 and/or components used to perform workflow steps may be implemented using computing resources of a provider network 190.
Col 10 line 17-27, A virtual compute instance may, for example, comprise one or more servers with a specified computational capacity (which may be specified by indicating the type and number of CPUs, the main memory size, and so on) and a specified software stack (e.g., a particular version of an operating system, which may in turn run on top of a hypervisor). A number of different types of computing devices may be used singly or in combination to implement the resources of the provider network in different embodiments, including general purpose or special purpose computer servers, storage devices, network devices, and the like.
col 11 line 22-32, The interactive workflow builder may determine additional prompt(s) to solicit configuration information for customizing a workflow template for the specific needs of the user. For example, the user interface may solicit input representing a training input dataset, references for a data collection step, data quality rules for a quality monitoring step, a specific model training algorithm for a training step, model approval rules for an approval step, locations at which to store outputs of various steps, a specific version of a model to be used for inference, and so on [based at least on the set of requirements ]
Col 11 line 33-45, As shown in 240, the method may determine and/or provision one or more computing resources to perform the generated workflow(s) [selecting]. In determining the resources, the machine learning management system may select or determine resource types, resource numbers, and resource configurations. The machine learning management system may provision the resources from one or more resource pools of a multi-tenant provider network. Provisioning a resource may include selecting the resource, reserving the resource for use by a particular account, configuring the resource to perform the desired task(s), and so on. The resource pools may include compute instances, storage resources, and other resources usable to perform machine learning tasks.
Col 11 line 51-67 and Col 12 line 1-4, As shown in 250, the steps of the workflow(s) may be performed using the selected computing resource(s). In one embodiment, the machine learning management system may generate a resource template that describes resources and their architecture for a particular workflow. The resource template may be provided to a cloud-based resource management service offered by the provider network [and a set of resource compatibility information characterizing capabilities of the plurality of resources relative to the set of requirements]. The resource template may be merged into a continuous deployment pipeline so that the workflow can be performed using the provisioned resources in the multi-tenant provider network. Workflows may be implemented using orchestration of various services of the provider network. For example, services in the provider network that implement machine learning tasks may include virtualized compute services that offer virtual compute instances, virtualized storage services that offer virtual storage resources, virtualized graphics processing services that offer virtual graphics processing units (GPUs), serverless computation services that execute code on behalf of clients, batch computing services that run batch computing jobs, machine learning endpoints in a machine learning framework, and so on [a combination of [[the]] available [[of]] resources of the distributed computing environment to perform the inferencing].);
[performing the inferencing using the] specified [machine learning model with the selected combination of the available resources of the distributed computing environment] (Yuan Col 6 line 35-44, The inference system 170 may include a plurality of endpoints. Each of endpoints may host one or more machine learning models that are used to generate inferences. Each of the endpoints may include one or more hosts or servers that perform inference tasks. The inference production may apply a trained machine learning model 155 to inference input data 146 in order to generate inferences 175. In one embodiment, the inferences 175 may be produced in substantially real-time, e.g., with minimal delays after the gathering of the inference input data 146.
Yuan col 11 line 22-32, The interactive workflow builder may determine additional prompt(s) to solicit configuration information for customizing a workflow template for the specific needs of the user. For example, the user interface may solicit input representing a training input dataset, references for a data collection step, data quality rules for a quality monitoring step, a specific model training algorithm for a training step, model approval rules for an approval step, locations at which to store outputs of various steps, a specific version of a model to be used for inference, and so on (i.e. specified machine learning model);
Horesh and Yuan are considered to be analogous to the claim invention because they are in the same field of machine learning. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filling date of the claimed invention to have modified Horesh to incorporate the teachings of Yuan to disclose determining a set a requirement and resource required for inferencing. Doing so to build and implement workflows for end-to-end machine learning solution and improving the quality of machine learning models (Yuan Col 2 line 37-52, Using the techniques described herein, a machine learning management system may manage an end-to-end machine learning model development lifecycle (MDLC) for rapid design, training, and productionization of high-quality machine learning models. The machine learning management system may implement workflows or blueprints for various stages of a lifecycle of a machine learning model, such as data sourcing, quality monitoring, feature engineering, model training, back-testing, evaluation and promotion, deployment (e.g., to produce inferences), and performance monitoring. By building and implementing workflows for an end-to-end machine learning lifecycle, the machine learning management system may accelerate the delivery of machine learning solutions while also improving the quality of machine learning models with versioning, auditing, and approval mechanisms.).
Regarding claim 2, Horesh and Yuan teach the method of claim 1.
Horesh further teaches wherein the combination of the available resources includes at least a machine learning engine and a machine learning accelerator (Horesh
Para 0028 line 1-5, After registration service 112 registers client device 120A, registration service 112 may request that model trainer 114 train a machine learning model (e.g., an inference engine 124) which registration service 112 can deploy to client device 120A [at least a machine learning engine].
Para 0106 line 8-16 Further, the various operations of methods described above may be performed by any suitable means capable of performing the corresponding functions. The means may include various hardware and/or software component(s) and/or module(s), including, but not limited to a circuit, an application specific integrated circuit (ASIC) [a machine learning accelerator], or processor.).
Regarding claim 3, Horesh and Yuan teach the method of claim 1.
Horesh further teaches further comprising: monitoring one or more available resources of the distributed computing environment to determine whether the one or more available resources are active and available to perform the inferencing before causing one or more new resource instances to be added to the plurality of resources (para 0084 line 1-4, At block 550, the central server broadcasts, to other client devices in the distributed computing system, an indication that the client device is available for participation in distributed inference generation.).
Regarding claim 4, Horesh and Yuan teach the method of claim 1.
Horesh teaches wherein the request is received from an application executing outside of the distributed computing environment in which the plurality of available resources is located (Horesh FIG. 1 – 110 (i.e. The “application executing from outside the distributed computing environment” is taught by the CENTRAL SERVER), 120A-C (i.e. The “distributed computing environment” is taught by the client devices 120A-C.)
PNG
media_image1.png
650
896
media_image1.png
Greyscale
).
Regarding claim 5, Horesh and Yuan teach the method of claim 4.
Horesh and Yuan are combine in the same rational as set forth above with respect to claim 1.
Horesh teaches wherein the distributed computing environment is a multi-tenant environment [in which at least a subset of the plurality of resources is able to be utilized by multiple users to perform different types of tasks] (Horesh Fig. 1,
PNG
media_image2.png
710
942
media_image2.png
Greyscale
[wherein the distributed computing environment is a multi-tenant environment]).
Yuan teaches [wherein the distributed computing environment is a multi-tenant environment] in which at least a subset of the plurality of resources is able to be utilized by multiple users to perform different types of tasks (Yuan Fig. 1,
PNG
media_image3.png
848
654
media_image3.png
Greyscale
Col 4 line 48-55, One or more users such as user 10 may be associated with a user account with a provider network 190. The user interface 30 may display or present a set of prompts or questions, and the user's answers to those questions may determine the presentation of additional prompts as well as the selection of one or more workflow templates and configuration of one or more workflows from the selected template(s).
Co 6l line 30-34, Any of the training 150 or inference 170 components may represent individual systems or subsystems that are loosely coupled or decoupled from one another. As shown in FIG. 1, the training 150 and inference 170 may be performed using resources of a multi-tenant provider network 190.
Col 7 line 35-41, The resource pools may include compute instances 194, storage resources 195, and other resources usable to perform machine learning tasks. To provision resources, the system 100 may interact with a resource manager 191 of the provider network 190 to select and reserve particular resources for use in performing workflows ( or portions thereof) on behalf of particular users. [at least a subset of the plurality of resources is able to be utilized by multiple users to perform different types of tasks].).
Regarding claim 6, Horesh and Yuan teach the method of claim 1.
Horesh further teaches further comprising: selecting the combination of the available resources based further upon at least resource configuration information or resource version information (Horesh para 0007 line 8-10, One or more client devices to be used in performing the inference are selected based, at least in part, on the confidence value for the local inference response [resource configuration information].).
Regarding claim 7, Horesh and Yuan teach the method of claim 1.
Horesh further teaches selecting the combination of the available resources based at least further upon one or more access control policies corresponding to the request, the specified machine learning model, or the available resources (Horesh Fig. 1
PNG
media_image4.png
688
856
media_image4.png
Greyscale
(i.e. Selecting the combination of resources based at least further upon […] the specified machine learning model” is taught by the determination of the number of peer devices based on the confidence value).
Regarding claim 8, Horesh and Yuan teach the method of claim 1.
Horesh further teaches, further comprising releasing the selected combination of the available resources after performing the inferencing (Horesh para 0032 line 7-12, In response , inference engine 124 at client device 120A may train ( or re - train ) the machine learning model using the training data set and indicate , to the registration service 112 , when training is complete and thus when the client device 120A is ready to accept inference requests from other client devices 120 in the computing environment 100.).
Regarding claim 9 and analogous claim 16, Horesh teaches identify, in a multi-tenant resource environment, a plurality of available resources available to perform the operation (Horesh Fig. 1,
PNG
media_image2.png
710
942
media_image2.png
Greyscale
para 0065 line 1-10, At block 220 , the system selects one or more client devices to be used in performing the inference with respect to the set of input data . The system can select one or more devices to be used in performing the inference based on a list of devices that are enrolled with a central server as candidate devices that can participate in a distributed inference generation procedure within a distributed computing environment . The system can select the one or more devices randomly , based on reputation scores , in a round robin fashion , or the like .);
and provide the (Horesh At block 240 , the system receives an inference response from each of the selected one or more client devices . The inference response received from each of the selected one or more client devices may include , for example , an inference generated by a client device (e.g. , a example , an inference generated by a client device ( e.g. , a category to which the input data set belongs ) and a confidence level associated with the inference.).
Horesh does not explicitly teach A system, comprising one or more processing units to receive a request to perform an operation using at least one artificial intelligence model; determine, based at least on one or more requirements characterizing one or more operational parameters of execution of the at least one artificial intelligence model to perform[[for]] the operation and a set of resource compatibility information characterizing capabilities of the plurality of available resources relative to the one or more requirements, a combination of the available resources to perform the
Yuan teaches A system, comprising: one or more processing units to: receive a request to perform an operation using at least one artificial intelligence model (Yuan Fig. 1,
PNG
media_image5.png
868
666
media_image5.png
Greyscale
col 11 line 22-32, The interactive workflow builder may determine additional prompt(s) to solicit configuration information for customizing a workflow template for the specific needs of the user. For example, the user interface may solicit input representing a training input dataset, references for a data collection step, data quality rules for a quality monitoring step, a specific model training algorithm for a training step, model approval rules for an approval step, locations at which to store outputs of various steps, a specific version of a model to be used for inference, and so on [an operation using at least one artificial intelligence model]);
determine, based at least on one or more requirements characterizing one or more operational parameters of execution of the at least one artificial intelligence model to perform[[for]] the operation and a set of resource compatibility information characterizing capabilities of the plurality of available resources relative to the one or more requirements, a combination of the available resources to perform the (Yuan Col 10 line 1-4, In one embodiment, components of the system 100 and/or components used to perform workflow steps may be implemented using computing resources of a provider network 190.
Col 10 line 17-27, A virtual compute instance may, for example, comprise one or more servers with a specified computational capacity (which may be specified by indicating the type and number of CPUs, the main memory size, and so on) and a specified software stack (e.g., a particular version of an operating system, which may in turn run on top of a hypervisor). A number of different types of computing devices may be used singly or in combination to implement the resources of the provider network in different embodiments, including general purpose or special purpose computer servers, storage devices, network devices, and the like.
col 11 line 22-32, The interactive workflow builder may determine additional prompt(s) to solicit configuration information for customizing a workflow template for the specific needs of the user. For example, the user interface may solicit input representing a training input dataset, references for a data collection step, data quality rules for a quality monitoring step, a specific model training algorithm for a training step, model approval rules for an approval step, locations at which to store outputs of various steps, a specific version of a model to be used for inference, and so on [determine,] [based at least on the set of requirements ]
Col 11 line 33-45, As shown in 240, the method may determine and/or provision one or more computing resources to perform the generated workflow(s) [selecting]. In determining the resources, the machine learning management system may select or determine resource types, resource numbers, and resource configurations. The machine learning management system may provision the resources from one or more resource pools of a multi-tenant provider network. Provisioning a resource may include selecting the resource, reserving the resource for use by a particular account, configuring the resource to perform the desired task(s), and so on. The resource pools may include compute instances, storage resources, and other resources usable to perform machine learning tasks.
Col 11 line 51-67 and Col 12 line 1-4, As shown in 250, the steps of the workflow(s) may be performed using the selected computing resource(s). In one embodiment, the machine learning management system may generate a resource template that describes resources and their architecture for a particular workflow. The resource template may be provided to a cloud-based resource management service offered by the provider network [characterizing one or more operational parameters of execution of the at least one artificial intelligence model to perform[[for]]]. The resource template may be merged into a continuous deployment pipeline so that the workflow can be performed using the provisioned resources in the multi-tenant provider network. Workflows may be implemented using orchestration of various services of the provider network. For example, services in the provider network that implement machine learning tasks may include virtualized compute services that offer virtual compute instances, virtualized storage services that offer virtual storage resources, virtualized graphics processing services that offer virtual graphics processing units (GPUs), serverless computation services that execute code on behalf of clients, batch computing services that run batch computing jobs, machine learning endpoints in a machine learning framework, and so on [the operation and a set of resource compatibility information characterizing capabilities of the plurality of available resources relative to the one or more requirements, a combination of the available resources to perform the ]);
Horesh and Yuan are combine in the same rational as set forth above with respect to claim 1.
Regarding claim 10 and claim 17, Horesh and Yuan disclose all the elements of claim 3 in system for rather than method form. Therefore, the supporting rationale of the rejection of claim 3 equally applies to claims 10 and 17.
Regarding claim 11 and claim 18, Horesh and Yuan disclose all the elements of claim 2 in system for rather than method form. Therefore, the supporting rationale of the rejection of claim 2 equally applies to claims 11 and 18.
Regarding claim 12 and claim 19, Horesh and Yuan disclose all the elements of claim 6 in system for rather than method form. Therefore, the supporting rationale of the rejection of claim 6 equally applies to claims 12 and 19.
Regarding claim 13 and claim 20, Horesh and Yuan disclose all the elements of claim 7 in system for rather than method form. Therefore, the supporting rationale of the rejection of claim 7 equally applies to claims 12 and 19.
Regarding claim 14, Horesh and Yuan teach the system of claim 9.
Horesh teaches wherein the system comprises at least one of: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system implemented using an edge device; a system implemented using a robot; a system for performing conversational AI operations; a system for generating synthetic data; a system incorporating one or more virtual machines (VMs);a system implemented at least partially in a data center; or a system implemented at least partially using cloud computing resources (Horesh para 0061 line 8-13, Inference engine 124 may anonymize the data included in the records retrieved from user data repository 140 prior to generating a training data set from the anonymized records and the synthetic data set and training or retraining the machine learning model based on the generated training data set [a system for generating synthetic data]).
Regarding claim 15, Horesh and Yuan disclose all the elements of claim 8 in system for rather than method form. Therefore, the supporting rationale of the rejection of claim 8 equally applies to claim 15.
Pertinent Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Kumar et al. (US20220150125A1) teaches a Cloud edge computing environment with cache models that have high frequency as shown in FIG. 3.
Alam et al. (US20220231964A1) – teaches a multi-tenant environment that allocates available space for job request (See figure 4 and 5).
Dean et al. (US20220357985A1) – teaches a resource manager to allocate portion of available computer resources such as hardware accelerators by appropriately load balancing.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALFREDO CAMPOS whose telephone number is (571)272-4504. The examiner can normally be reached 7:00 - 4:00 pm M - F.
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, Michael J. Huntley can be reached at (303) 297-4307. 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.
/ALFREDO CAMPOS/Examiner, Art Unit 2129
/MICHAEL J HUNTLEY/Supervisory Patent Examiner, Art Unit 2129
Prosecution Insights