DETAILED ACTION
This action is in response to application 18/635951, filed on 4/15/2024. Claims 1-20 are pending. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
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 (i.e., changing from AIA to pre-AIA ) 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, 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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Chen et al., “Foundation Model Based Native AI Framework in 6G with Cloud-Edge-End Collaboration,” arXiv, published 26 October 2023, hereinafter “Chen,” and Zawish et al., “Complexity-Driven Model Compression for Resource-Constrained Deep Learning on Edge,” IEEE, published 12 January 2024, hereinafter “Zawish.”
Regarding claim 1, Chen discloses “A computer-implemented method (CIM), comprising:
receiving a request for a new AI based model deployment; (see, e.g., Chen, pg. 3, sec. III, “we propose an intelligent synergistic cloud edge-end framework, depicted in Fig. 1. The cloud identifies user intent by processing queries from the end-user device using the intent-aware PFM. It then manages task orchestration and resource allocation for the edge or end-user device by utilizing algorithms within the edge algorithm toolkit.”)
determining a combination of resources that are configured to satisfy the received request; (see, e.g., Chen, pg. 3, sec. III, “we begin by discussing the deployment and fine tuning of the native intent-aware PFM in the cloud, which is the core of the native AI framework.”; pg. 5, Fig. 3; pg. 6, sec. III(D); “Within the cloud, the intent aware PFM is implemented, facilitating intent recognition based on the uploaded information. The PFM orchestrates the tasks and encapsulates the interrelationships between sub-tasks as a directed acyclic graph (DAG), which is then fed back to the edge. Following the DAG, the edge node selects the appropriate models from its toolkit and then either conducts the task with the end-user device’s data or sends the models to the end-user device for local execution.”)
in response to determining that at least one of the resources in the combination of resources is unavailable, determining whether to re-configure resources used to form one or more existing AI based model deployments to satisfy the received request; (see, e.g., Chen, pg. 2, sec. II(B); “deploying an entire PFM on a resource-constrained device is often unfeasible, leading to the need for task-oriented AI models. However, despite enormous efforts spent in enhancing AI in 6G, the industry still lacks a standardized library of task-oriented AI models to address practical challenges.”; “Given the significant computational resource demands associated with training PFMs, deploying them at the network edge or on end-user devices is often impractical. Instead, the use of lightweight AI models aligns with the vision of achieving ubiquitous intelligence in 6G.”)
Chen does not independently disclose the further limitations listed below. More precisely, while Chen clearly describes resource deficiencies, Chen does not explicitly disclose reconfiguring (e.g., compressing, as in claim 2) AI model deployments to overcome resource deficiencies at edge and end-user devices. Zawish discloses that while “CNNs deployed at Internet of Things (IoT) edges can enable pervasive smart services characterized by exceptional efficiency and strict privacy standards . . . CNNs cannot be ported on resource-constrained edge devices because of their highly parameterized nature” and “[a]s a result, CNNs need to be compressed to reduce resource demands for inference on edge.” (Zawish pg. 3886, sec. I). Zawish further discloses “structured pruning at initialization, which not only accelerates CNN inference but also improves training efficiency on edge.” (Zawish, pg. 3887).
Chen and Zawish are directed toward AI computation and therefore are analogous art. On or before the effective filing date of the instant claim, one of ordinary skill in the art would have deemed it obvious to try to combine the model compression of Zawish with the edge model deployment of Chen, thereby obtaining the limitations:
causing the resources used to form the one or more existing AI based model deployments to be re-configured; (see, e.g., Zawish, pg. 3892; “reduction in memory of a CNN is critical . . . in order for the deep models to run at the edge, they must fit within the target device’s RAM without disrupting the IoT application at the runtime.”) and
re-deploying the re-configured resources, by: causing updated versions of the one or more existing AI based model deployments to be formed, (see, e.g., Zawish, pg. 3892) and causing the requested new AI based model deployment to be formed.” (see, e.g., Chen, pg. 4, sec. III(A); “to achieve superior performance with a native PFM, another approach involves a direct full model update on the foundation model”).
A clear and predictable benefit of combining Chen and Zawish would have appeared as the new ability, afforded by reconfigured compressed AI models, of “deploying them at the network edge or on end-user devices.” (Chen, pg. 2, sec. II(B)). Accordingly, the instant claim is unpatentable over the combination of Chen and Zawish.
Regarding claim 2, the combination of Chen and Zawish renders obvious “The CIM of claim 1, wherein the updated versions of the one or more existing AI based model deployments are formed by: compressing the resources used to form the one or more existing AI based model deployments; and compressing the combination of resources configured to satisfy the received request.” (see, e.g., Zawish, pg. 3886; “CNNs need to be compressed to reduce resource demands for inference on edge.”).
Regarding claim 3, the combination of Chen and Zawish renders obvious “The CIM of claim 1, further comprising: monitoring current states of the existing AI based model deployments by: gathering information associated with the current states of the existing AI based model deployments; and using the information to perform capacity profiling.” (see, e.g., Chen, pg. 4, sec. III(B); “The intelligent edge within this framework pools the majority of its local resources. These resources are represented as part of the edge’s status information, which is then transmitted to the cloud along with requests from end-user devices.”; pg. 5, sec. III(B); “AI models stored in the algorithm toolkits are orchestrated by the well trained PFM to handle tasks across multiple edge/end-user devices. Through intent recognition and unified orchestration, these models are assigned to tasks that align with their capabilities, enabling them to effectively leverage the relationships between tasks.”).
Regarding claim 4, the combination of Chen and Zawish renders obvious “The CIM of claim 3, wherein the information associated with the current states of the existing AI based model deployments is selected from the group consisting of: priority information, performance bounds, and resource utilizations.” (see, e.g., Chen, pg. 4, sec. III(B); “resource scheduling and task orchestration for a cell have shifted from its edge server to the cloud-based intent aware PFM within this framework. As a result, each intelligent edge node is now an integral part of the whole system, rather than functioning as separate service nodes. This shift allows them to establish task-oriented connections among various intelligent edge nodes through cloud resource allocation and task orchestration. This form of connection enables intelligent edge nodes to share computational resources and offload computational tasks.”).
Regarding claim 5, the combination of Chen and Zawish renders obvious “The CIM of claim 3, wherein the using of the information to perform capacity profiling includes: outputting a prioritized list of the existing AI based model deployments and their respective current states.” (see, e.g., Chen, pg. 5, Fig. 3 depicts existing AI based model deployments and their respective states and priority levels).
Regarding claim 6, the combination of Chen and Zawish renders obvious “The CIM of claim 1, wherein the causing of the resources used to form the one or more existing AI based model deployments to be re-configured includes:
causing the one or more existing AI based model deployments to be stored in memory; (see, e.g., Chen, pg. 5, Fig. 3 depicts storing AI models)
causing the one or more existing AI based model deployments to be unloaded; (see, e.g., Chen, pg. 5, Fig. 3 depicts unloading and loading AI models) and
re-configuring the resources from the one or more unloaded AI based model deployments to form: the updated versions of the one or more existing AI based model deployments, (see, e.g., Zawish, pg. 3886-3887) and the requested new AI based model deployment.” (see, e.g., Chen, pg. 4, sec. III(A); “to achieve superior performance with a native PFM, another approach involves a direct full model update on the foundation model”).
Regarding claim 7, the combination of Chen and Zawish renders obvious “The CIM of claim 6, wherein the causing of the one or more existing AI based model deployments to be stored in memory includes: storing running parameters of the one or more existing AI based model deployments.” (see, e.g., Chen, pg. 5, Fig. 3 depicts storing AI models and parameters).
Regarding claim 8, the combination of Chen and Zawish renders obvious “The CIM of claim 1, wherein one or more of the existing AI based model deployments include multi-objective foundation models on at least one edge device.” (See, e.g., Chen, pg. 1, sec. I; “we establish a 6G native AI framework based on a pre-trained foundation model (PFM).”).
Regarding claim 9, the combination of Chen and Zawish renders obvious “The CIM of claim 8, wherein the requested new AI based model deployment includes one or more multi-objective foundation models.” (see, e.g., Chen, pg. 4, sec. III(A); “to achieve superior performance with a native PFM, another approach involves a direct full model update on the foundation model”).
Regarding claim 10, the combination of Chen and Zawish renders obvious “The CIM of claim 8, wherein the operations are performed by a central server connected to the at least one edge device.” (See, e.g., Chen, pg. 4, sec. III(B); “resource scheduling and task orchestration for a cell have shifted from its edge server to the cloud-based intent aware PFM within this framework.”).
Regarding claims 11-20, the instant claims are equivalents of claims 1-10, differing only by statutory class. Accordingly, the rejection of claim 1 applies, mutatis mutandis, to claims 11 and 20; the rejection of claim 2 applies, mutatis mutandis, to claim 12; the rejection of claim 3 applies, mutatis mutandis, to claim 13; the rejection of claim 4 applies, mutatis mutandis, to claim 14; the rejection of claim 5 applies, mutatis mutandis, to claim 15; the rejection of claim 6 applies, mutatis mutandis, to claim 16; the rejection of claim 7 applies, mutatis mutandis, to claim 17; the rejection of claim 8 applies, mutatis mutandis, to claim 18; and the rejections of claims 9-10 apply, mutatis mutandis, to claim 19.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RYAN D COYER whose telephone number is (571) 270-5306. The examiner can normally be reached Monday-Friday 12pm-10pm Eastern Time. 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, Wei Mui, can be reached on 571-272-3708. 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.
/Ryan D. Coyer/Primary Examiner, Art Unit 2191
Prosecution Insights