DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
2. Claims 1–20 are presented for examination in a non-provisional application filed on 01/23/2024.
Drawings
3. The drawings were received on 01/23/2024 (in the filings). These drawings are acceptable.
Double Patenting
4. The nonstatutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) so as to prevent the unjustified or improper timewise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A nonstatutory double patenting rejection is appropriate where the claims at issue are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); and In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on a nonstatutory double patenting ground provided the reference application or patent either is shown to be commonly owned with this application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b).
The USPTO internet Web site contains terminal disclaimer forms which may be used. Please visit http://www.uspto.gov/forms/. The filing date of the application will determine what form should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to http://www.uspto.gov/patents/process/file/efs/guidance/eTD-info-I.jsp.
A.
5. Claims 1–20 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1–20 of Copending Application No. 18/419,757 (Pub. No. US 2025/0238710 A1), further in view of (A) Olgiati et al., US 2025/0173597 A1 (“Olgiati”), as applied in rejecting claims 1–20 below.
6. Although the claims at issue are not identical, they are not patentably distinct (nonobvious) from each other, because at least some of the subject matter claimed in the instant application is already fully disclosed in the Copending Application No. 18/419,757.
This is a provisional nonstatutory double patenting rejection because the patentably indistinct claims have not in fact been patented.
For purposes of illustration, a table has been constructed below to compare the two independent method claims and one or more exemplary dependent claims.
Instant Application No. 18/419,780
Copending Application No. 18/419,757
1. A method for managing workload placement, the method comprising:
obtaining, by a workload placement service, a request for assigning a model adaptation workload to one of a plurality of production environments based on latency minimization;
in response to the request:
performing an initial workload placement to assign the model adaptation workload to a first production environment of the plurality of production environments;
after performing the initial workload placement, monitoring:
execution of the model adaptation workload on the first production environment, and
communication between the first production environment and a second production environment executing a corresponding inferencing workload, to obtain telemetry data associated with the execution and the communication;
performing a latency analysis using the telemetry data to generate a placement recommendation;
making a determination that the placement recommendation specifies a third production environment of the plurality of production environments; and
based on the determination, initiating deployment of the model adaptation workload to the third production environment.
1. A method for managing workload placement, the method comprising:
obtaining, by a workload placement service, a request for assigning a training workload to one of a plurality of production environments based on completion time;
in response to the request:
performing an initial workload placement of the training workload to assign the training workload to a first production environment of the plurality of production environments;
after performing the initial workload placement, monitoring:
execution of the training workload on the first production environment, and
performance of computing resource in the plurality of production environments to obtain telemetry data associated with the execution and the performance;
performing a completion time analysis using the telemetry data to generate a placement recommendation;
making a determination that the placement recommendation specifies a second production environment of the plurality of production environments; and
based on the determination, initiating deployment of the training workload to the second production environment.
2. The method of claim 1, wherein the model adaptation workload comprises performing a parameter-efficient fine-tuning (PEFT) process on the inferencing workload, and wherein the inferencing workload comprises an implementation of a generative artificial intelligence (AI) model.
2. The method of claim 1, wherein the training workload comprises the training of a generative artificial intelligence (AI) model using training data.
3. The method of claim 2, wherein a front-end device is operated by a user and utilizes the generative AI model to obtain an inferencing payload.
3. The method of claim 2, wherein the generative AI model is utilized by a front-end environment to obtain an inferencing payload.
The Examiner notes that Copending Application No. 18/419,757 is directed to workload placement “based on completion times” the optimization of which requires or renders obvious maximizing processing performance of hosts and/or minimizing latencies between hosts.
Moreover, Olgiati teaches or suggests “wherein the model adaptation workload comprises performing a parameter-efficient fine-tuning (PEFT) process on the inferencing workload, and wherein the inferencing workload comprises an implementation of a generative artificial intelligence (AI) model” (¶¶ 24 and 71, as applied in rejecting claim 2 below). It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Olgiati to so as to fine-tuning the inferencing models.
B.
7. Claims 1–20 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1–20 of Copending Application No. 18/419,666 (Pub. No. US 2025/0238274 A1), further in view of (A) Olgiati et al., US 2025/0173597 A1 (“Olgiati”), as applied in rejecting claims 1–20 below.
8. Although the claims at issue are not identical, they are not patentably distinct (nonobvious) from each other, because at least some of the subject matter claimed in the instant application is already fully disclosed in the Copending Application No. 18/419,666.
This is a provisional nonstatutory double patenting rejection because the patentably indistinct claims have not in fact been patented.
For purposes of illustration, a table has been constructed below to compare the two independent method claims and one or more exemplary dependent claims.
Instant Application No. 18/419,780
Copending Application No. 18/419,666
1. A method for managing workload placement, the method comprising:
obtaining, by a workload placement service, a request for assigning a model adaptation workload to one of a plurality of production environments based on latency minimization;
in response to the request:
performing an initial workload placement to assign the model adaptation workload to a first production environment of the plurality of production environments;
after performing the initial workload placement, monitoring:
execution of the model adaptation workload on the first production environment, and
communication between the first production environment and a second production environment executing a corresponding inferencing workload, to obtain telemetry data associated with the execution and the communication;
performing a latency analysis using the telemetry data to generate a placement recommendation;
making a determination that the placement recommendation specifies a third production environment of the plurality of production environments; and
based on the determination, initiating deployment of the model adaptation workload to the third production environment.
1. A method for managing workload placement, the method comprising:
obtaining, by a workload placement service, a request for assigning an inferencing workload to one of a plurality of production environments based on latency minimization;
in response to the request:
performing an initial workload placement of the inferencing workload to assign the inferencing workload to a first production environment of the plurality of production environments;
after performing the initial workload placement, monitoring:
execution of the inferencing workload on the first production environment, and
communication between the first production environment and a front-end environment, to obtain telemetry data associated with the execution and the communication;
performing a latency analysis using the telemetry data to generate a placement recommendation;
making a determination that the placement recommendation specifies a second production environment of the plurality of production environments; and
based on the determination, initiating deployment of the inferencing workload to the second production environment.
2. The method of claim 1, wherein the model adaptation workload comprises performing a parameter-efficient fine-tuning (PEFT) process on the inferencing workload, and wherein the inferencing workload comprises an implementation of a generative artificial intelligence (AI) model.
2. The method of claim 1, wherein the inferencing workload comprises implementing a generative artificial intelligence (AI) model.
3. The method of claim 2, wherein a front-end device is operated by a user and utilizes the generative AI model to obtain an inferencing payload.
3. The method of claim 2, wherein the front-end environment comprises a front-end device operated by a user utilizing the generative AI model to obtain an inferencing payload.
The Examiner notes that Copending Application No. 18/419,666 is directed to placement of the “inferencing” workload. It would have been obvious to a person of ordinary skill in the art to optimize the placement (i.e. location) of either types of workloads to minimize the latency between the placement hosts, e.g. moving either the adaptation or inferencing workload.
Moreover, Olgiati teaches or suggests “wherein the model adaptation workload comprises performing a parameter-efficient fine-tuning (PEFT) process on the inferencing workload, and wherein the inferencing workload comprises an implementation of a generative artificial intelligence (AI) model” (¶¶ 24 and 71, as applied in rejecting claim 2 below). It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of Olgiati to so as to fine-tuning the inferencing models.
C.
9. Claims 1–20 are rejected on the ground of nonstatutory double patenting as being unpatentable over claims 1–20 of Copending Application No. 18/419,790 (Pub. No. US 2025/0238275 A1).
10. Although the claims at issue are not identical, they are not patentably distinct (nonobvious) from each other, because at least some of the subject matter claimed in the instant application is already fully disclosed in the Copending Application No. 18/419,790.
This is a provisional nonstatutory double patenting rejection because the patentably indistinct claims have not in fact been patented.
For purposes of illustration, a table has been constructed below to compare the two independent method claims and one or more exemplary dependent claims.
Instant Application No. 18/419,780
Copending Application No. 18/419,790
1. A method for managing workload placement, the method comprising:
obtaining, by a workload placement service, a request for assigning a model adaptation workload to one of a plurality of production environments based on latency minimization;
in response to the request:
performing an initial workload placement to assign the model adaptation workload to a first production environment of the plurality of production environments;
after performing the initial workload placement, monitoring:
execution of the model adaptation workload on the first production environment, and
communication between the first production environment and a second production environment executing a corresponding inferencing workload, to obtain telemetry data associated with the execution and the communication;
performing a latency analysis using the telemetry data to generate a placement recommendation;
making a determination that the placement recommendation specifies a third production environment of the plurality of production environments; and
based on the determination, initiating deployment of the model adaptation workload to the third production environment.
1. A method for managing workload placement, the method comprising:
obtaining, by a workload placement service, a request for assigning a model adaptation workload to one of a plurality of production environments based on completion time;
in response to the request:
performing an initial workload placement to assign the model adaptation workload to a first production environment of the plurality of production environments;
after performing the initial workload placement, monitoring:
execution of the model adaptation workload on the first production environment, and
performance of computing resource in the plurality of production environments to obtain telemetry data associated with the execution and the performance;
performing a completion time analysis using the telemetry data to generate a placement recommendation;
making a determination that the placement recommendation specifies a second production environment of the plurality of production environments; and
based on the determination, initiating deployment of the model adaptation workload to the second production environment
.
2. The method of claim 1, wherein the model adaptation workload comprises performing a parameter-efficient fine-tuning (PEFT) process on the inferencing workload, and wherein the inferencing workload comprises an implementation of a generative artificial intelligence (AI) model.
2. The method of claim 1, wherein the model adaptation workload comprises performing a parameter-efficient fine-tuning (PEFT) process on a training workload, and wherein the training workload comprises a training of a generative artificial intelligence (AI) model using a training dataset.
3. The method of claim 2, wherein a front-end device is operated by a user and utilizes the generative AI model to obtain an inferencing payload.
3. The method of claim 2, wherein the generative AI model is utilized by a front-end environment to obtain an inferencing payload.
The Examiner notes that Copending Application No. 18/419,790 is directed to workload placement “based on completion times” the optimization of which requires or renders obvious maximizing processing performance of hosts and/or minimizing latencies between hosts.
Examiner’s Remarks
11. Examiner refers to and explicitly cites particular pages, sections, figures, paragraphs or columns and lines in the references as applied to Applicant’s claims to the extent practicable to streamline prosecution.
Although the cited portions of the references are representative of the best teachings in the art and are applied to meet the specific limitations of the claims, other uncited but related teachings of the references may be equally applicable as well. It is respectfully requested that, in preparing responses to the rejections, the Applicant fully considers not only the cited portions of the references, but also the references in their entirety, as potentially teaching, suggesting or rendering obvious all or one or more aspects of the claimed invention.
Abbreviations
12. Where appropriate, the following abbreviations will be used when referencing Applicant’s submissions and specific teachings of the reference(s):
i. figure / figures: Fig. / Figs.
ii. column / columns: Col. / Cols.
iii. page / pages: p. / pp.
References Cited
13. (A) Olgiati et al., US 2025/0173597 A1 (“Olgiati”).
(B) Toledo et al., US 2018/0331905 A1 (“Toledo”).
Notice re prior art available under both pre-AIA and AIA
14. 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.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
A.
15. Claims 1–20 are rejected under 35 U.S.C. 103 as being unpatentable over (A) Olgiati in view of (B) Toledo.
See “References Cited” section, above, for full citations of references.
16. Regarding claim 1, (A) Olgiati teaches/suggests the invention substantially as claimed, including:
“A method for managing workload placement, the method comprising:
obtaining, by a workload placement service, a request for assigning a model adaptation workload to one of a plurality of production environments based on latency minimization;
(¶¶ 97–98: 1410, a request to place a machine learning model on a host system of the machine learning service may be received. The machine learning model may be a base model for a finetuned machine learning model, in some embodiments.
As indicated at 1420, different machine learning models that are respective delta models with respect to the base model may be identified, where respective combinations of the delta models with the base model produce respective versions of the fine-tuned machine learning model;
¶ 99: both the base model and the respective delta models may be placed on the host system, the host system generates respective inferences for requests that invoke one of the respective versions of the fine-tuned machine learning model;
¶ 68: ADAPT the trained machine learning model's performance to specific uses or scenarios included in the additional training data;
¶ 69: machine learning models can be extremely large using, for instance, billions of parameters, allowing the model to be adaptable to a wide category of use cases and tasks, such as text and image generation and summarization. These machine learning models, which are sometimes referred to as “foundation models”, may perform well without any adaptation. However, in many scenarios, better performance can be achieved if the models are fine-tuned to specific uses cases;
¶ 27: In various embodiments, optimal placement strategies, including optimized placement of fine-tuned machine learning models at host systems, may be implemented to improve inference performance by co-locating related machine learning models together to avoid various latency penalties;
¶ 23: The training and validation process may be repeated periodically or intermittently, by using new training data to refine previously learned parameters of a production model and deploy a new production model for inference);
in response to the request:
performing an initial workload placement to assign the model adaptation workload to a first production environment of the plurality of production environments;
(¶ 99: both the base model and the respective delta models may be placed on the host system, the host system generates respective inferences for requests that invoke one of the respective versions of the fine-tuned machine learning model)
after performing the initial workload placement, monitoring:
execution of the model adaptation workload on the first production environment”
(¶ 58: monitoring a managed network endpoint for dynamic endpoint management for heterogeneous machine learning models, according to some embodiments. Endpoint monitoring 219 may be implemented as part of endpoint management 215 to proactively address potential failure or other performance problems and maintain or improve performance;
¶ 59: Replica/instance metrics 502 may report various performance and utilization measures for individual replicas of a model and their respective host instances. Some metrics may include various computing resource utilization metrics, such as CPU utilization, GPU utilization, memory utilization, reservations, disk or other storage utilization, and inference performance metrics;
¶ 60: Endpoint monitoring 219 may implement model replica rebalancing 217, which may examine the placement and performance of model replicas 533a and 535a in inference
containers 532a and 534a across host instances 530a to ensure efficient utilization of hosts 530a ...
performing a latency analysis using the telemetry data to generate a placement recommendation
(¶ 59: replica latency;
¶ 60: Other instances of unhealthy or poor placement may be indicated by performance metric such as number of errors or LATENCY OF INFERENCES. Thus, rebalancing events may be triggered for performance when various criteria (e.g., thresholds) analyzing these metrics are satisfied;
¶ 61: model placement 216 may return possible placements which model replica rebalancing 217 may confirm before initiating (as indicated at 506). In this way, model replica rebalancing 217 can determine whether a placement improves the situation)
making a determination that the placement recommendation specifies a third production environment of the plurality of production environments; and
(¶ 61: model placement 216 may return possible placements which model replica rebalancing 217 may confirm before initiating (as indicated at 506). In this way, model replica rebalancing 217 can determine whether a placement improves the situation ...
In some scenarios, rebalancing events may NOT be performed due to lack of an alternative placement location
the Examiner notes: if no alternative location is identified / specified then placement would not be performed, which requires or renders obvious determining availability of possible locations (recommendation) before initiating rebalancing);
based on the determination, initiating deployment of the model adaptation workload to the third production environment”
(¶ 61: To handle these detected rebalancing events, model replica rebalancing 217 may get replica placements 504 to move replicas).
Olgiati do not explicitly teach “monitoring ... communication between the first production environment and a second production environment executing a corresponding inferencing workload, to obtain telemetry data associated with the execution and the communication.”
(B) Toledo, in the context of Olgiati’s teachings, however teaches or suggests implementing:
“monitoring ... communication between the first production environment and a second production environment executing a corresponding inferencing workload, to obtain telemetry data associated with the execution and the communication”
(¶ 51: The processing module 202 is configured to monitor the traffic flows for a predefined sampling period. The predefined sampling period may be chosen to be any length of time based on user input or in some cases, the processing module 202 may select the predefined sampling period based on application usage patterns;
¶ 52: Each traffic flow includes one or more flow segments, such as for example, in the above example, traffic flow between microservice instances S1, S2 and S3 involve flow segments corresponding to flow of communication between S1 and S2, S2 and S3, S3 and S2, and S2 and S1;
¶ 77: Each host is further depicted to include a latency probe. For example, host H1 is depicted to include a latency probe 406a; host H2 is depicted to include a latency probe 406b; and host H3 is depicted to include a latency probe 406c. The links 414a, 414b, 416a and 416b depict latency measured between the hosts H1 and H2;
¶ 78: volume of all traffic associated with each traffic flow in a day or in one-hour time period may be monitored. Moreover, the processing module 202 also measures baseline latencies across hosts in milliseconds);
¶ 54: identifies a pair of microservice instances associated with the flow segment. In an illustrative example, the processing module 202 may first determine if at least one microservice instance from among the pair of microservice instances is capable of being relocated from a respective current host to another host for achieving at least one predetermined performance objective. Some non-limiting examples of the predetermined performance objective may include an objective related to reducing a volume of communication between hosts deploying the pair of microservice instances, an objective related to REDUCING A COMMUNICATION LATENCY between the hosts deploying the pair of microservice instances, and the like);
¶ 57: determines if the pair of microservice instances can be relocated to a host pair with least latency. To that effect, the processing module 202 may be configured to identify another host associated with least baseline latency from among several hosts for relocating the at least one microservice instance to achieve the predetermined performance objective. It is noted that the identified host for relocating the microservice instance may be chosen for relocation only upon successful evaluation of the predefined criteria);
It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine the teachings of (B) Toledo with those of (A) Olgiati to monitor/probe communications between base model and the respective delta models for generating respective inferences. The motivation or advantage to do so is to improve the performance of inferencing models based on accurate, real-time latency measurements.
17. Regarding claim 2, Olgiati teaches or suggests:
“wherein the model adaptation workload comprises performing a parameter-efficient fine-tuning (PEFT) process on the inferencing workload, and wherein the inferencing workload comprises an implementation of a generative artificial intelligence (AI) model”
(¶ 71: Parameter efficient fine-tuning refers to a set of fine-tuning techniques that do not require updating all the model weights;
¶ 24: generative machine learning models (sometimes referred to as generative artificial intelligence (AI)) are being integrated into machine learning (ML) applications to support performance of various tasks).
18. Regarding claim 3, Olgiati teaches or suggests:
“wherein a front-end device is operated by a user and utilizes the generative AI model to obtain an inferencing payload”
(¶ 41: control plane 212 may include components that support the management of different types of endpoints, both network endpoint(s) 224a, which may be manually managed by a user of machine
¶ 47: clients 250 may encompass any type of client that can submit network-based requests to
provider network 200 via network 260, including requests for machine learning service 210 (e.g., a request to start machine learning task execution, etc.;
¶ 48: client 250 may provide access to provider network 200 to other applications in a manner that is transparent to those applications. Clients 250 may convey network-based services requests (e.g., access requests to configure or perform machine learning tasks) via network 260;
¶¶ 97–98: 1410, a request to place a machine learning model on a host system of the machine learning service may be received. The machine learning model may be a base model for a finetuned machine learning model, in some embodiments.
As indicated at 1420, different machine learning models that are respective delta models with respect to the base model may be identified, where respective combinations of the delta models with the base model produce respective versions of the fine-tuned machine learning model;
¶ 99: both the base model and the respective delta models may be placed on the host system, the host system generates respective inferences for requests that invoke one of the respective versions of the fine-tuned machine learning model).
19. Regarding claim 4, Olgiati and Toledo, in combination, teach or suggest:
“wherein the latency analysis is further based on causal variables associated with latency in the communication”
(Olgiati — ¶ 59: replica latency;
¶ 60: Other instances of unhealthy or poor placement may be indicated by performance metric such as number of errors or LATENCY OF INFERENCES. Thus, rebalancing events may be triggered for performance when various criteria (e.g., thresholds) analyzing these metrics are satisfied;
Toledo — ¶ 77: Each host is further depicted to include a latency probe. For example, host H1 is depicted to include a latency probe 406a; host H2 is depicted to include a latency probe 406b; and host H3 is depicted to include a latency probe 406c. The links 414a, 414b, 416a and 416b depict latency measured between the hosts H1 and H2;
¶ 78: volume of all traffic associated with each traffic flow in a day or in one-hour time period may be monitored. Moreover, the processing module 202 also measures baseline latencies across hosts in milliseconds).
20. Regarding claim 5, Olgiati and Toledo, in combination, teach or suggest:
“wherein the causal variables comprise at least one of: ... latency between GPUs executing the model adaptation workload”
(Olgiati — ¶ 59: replica latency;
¶ 60: Other instances of unhealthy or poor placement may be indicated by performance metric such as number of errors or LATENCY OF INFERENCES. Thus, rebalancing events may be triggered for performance when various criteria (e.g., thresholds) analyzing these metrics are satisfied;
¶ 26: Varying host systems with different hardware or other performance capabilities, such as hosts optimized for generative AI with multi-GPU;
¶ 44: GPUs;
¶¶ 97–98: 1410, a request to place a machine learning model on a host system of the machine learning service may be received. The machine learning model may be a base model for a finetuned machine learning model, in some embodiments.
As indicated at 1420, different machine learning models that are respective delta models with respect to the base model may be identified, where respective combinations of the delta models with the base model produce respective versions of the fine-tuned machine learning model;
¶ 99: both the base model and the respective delta models may be placed on the host system, the host system generates respective inferences for requests that invoke one of the respective versions of the fine-tuned machine learning model;
Toledo — ¶ 77: Each host is further depicted to include a latency probe. For example, host H1 is depicted to include a latency probe 406a; host H2 is depicted to include a latency probe 406b; and host H3 is depicted to include a latency probe 406c. The links 414a, 414b, 416a and 416b depict latency measured between the hosts H1 and H2;
¶ 78: volume of all traffic associated with each traffic flow in a day or in one-hour time period may be monitored. Moreover, the processing module 202 also measures baseline latencies across hosts in milliseconds).
21. Regarding claim 6, Olgiati teaches or suggests:
“wherein the first production environment is a computing device of an on-premise environment”
(¶ 33: Provider network 200 may be a private or closed system;
¶ 35: an edge location can be a data center positioned to provide capacity to a set of customers within a certain latency requirement, a set of servers provided to a customer's premises;
¶ 48: both a given client 250 and provider network 200 may be respectively provisioned within enterprises having their own internal networks).
22. Regarding claim 7, Olgiati teaches or suggests:
“wherein the first production environment is a computing device of a cloud environment operatively connected to a front-end environment via a wide area network”
(¶ 33: Provider network 200 ... may be set up by an entity such as a company or a public sector organization to provide one or more services (such as various types of cloud-based storage) accessible via the Internet and/or other networks to clients 250, in one embodiment;
¶ 34: provider network 200 can be formed as a number of regions, where a region is a separate geographical area in which the cloud provider clusters data centers;
¶ 41: control plane 212 may include components that support the management of different types of endpoints, both network endpoint(s) 224a, which may be manually managed by a user of machine
¶ 47: clients 250 may encompass any type of client that can submit network-based requests to provider network 200 via network 260, including requests for machine learning service 210 (e.g., a request to start machine learning task execution, etc.;
¶ 48: wide area networks;
¶¶ 97–98: 1410, a request to place a machine learning model on a host system of the machine learning service may be received. The machine learning model may be a base model for a finetuned machine learning model, in some embodiments.
As indicated at 1420, different machine learning models that are respective delta models with respect to the base model may be identified, where respective combinations of the delta models with the base model produce respective versions of the fine-tuned machine learning model;
¶ 99: both the base model and the respective delta models may be placed on the host system, the host system generates respective inferences for requests that invoke one of the respective versions of the fine-tuned machine learning model).
23. Regarding claims 8–14, they are the corresponding computer program product claims reciting similar limitations of commensurate scope as the method of claims 1–7, respectively. Therefore, they are rejected on the same basis as claims 1–7 above.
24. Regarding claims 15–20, they are the corresponding system claims reciting similar limitations of commensurate scope as the method of claims 1–3 and 5–7, respectively. Therefore, they are rejected on the same basis as claims 1–3 and 5–7 above, including the following rationale:
Olgiati teaches or suggests “processor; and memory including instructions, which when executed by the processor, perform a method”
(Fig. 19 and ¶ 115: processors executing program instructions stored on one or more computer-readable storage media coupled to the processors).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
(a) Li et al., US 2024/0054366 A1, teaching artificial intelligence (AI) application deployment method including a development system converts a trained AI model into at least one adaptation model.
(b) Rao et al., US 2024/0403137 A1, teaching dynamically optimizing microservice placement in a distributed edge and cloud computing environment.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BENJAMIN C WU whose telephone number is (571)270-5906. The examiner can normally be reached Monday through Friday, 8:30 A.M. to 5:00 P.M..
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, Aimee J. Li can be reached on (571)272-4169. 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.
/BENJAMIN C WU/Primary Examiner, Art Unit 2195
June 4, 2026
Prosecution Insights