DETAILED ACTION
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 the application filed 12/05/2023.
Claims 1-20 are presented for examination. Claims 1, 14, and 20 are independent Claims.
Drawings
2. The drawings filed 12/05/2023 are acceptable for examination purposes.
Claim Rejections - 35 USC § 102
3. 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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by ZHANG et al. (US 20230334320).
As to Claim 1:
ZHANG teaches a system (Abstract), comprising:
a memory (Fig. 17); and
a processor (Fig. 17), wherein the processor is configured to:
search, by utilizing a neural network, for a plurality of modules for inclusion in a module collection of a search space ([0002-0003]);
determine, by utilizing the neural network, an insertion point within an existing artificial intelligence model ([0034-0035]);
apply a metric to the plurality of modules in the module collection ([0035-0037]);
generate, based on the metric, a ranking of candidate modules of the plurality of modules for substituting an existing module located at the insertion point within the existing artificial intelligence model ([0035-0037] and [0040-0042]);
determine an accuracy rank for each of the candidate modules ([0003], [0037], [0039], [0059], and [0122]);
facilitate execution of candidate models including the candidate modules on a deep learning accelerator to determine a runtime execution rank for each of the candidate models ([0120-0124]); and
determine an optimal proposed model from the candidate models based on the accuracy rank and the runtime execution rank ([0071-0072] and [0099-0102]).
As to Claim 2:
ZHANG teaches train the candidate modules using intermediate features distillation over a period of time on a distribution of data associated with a dataset; determine the accuracy rank for each of the candidate modules based on the training of the candidate modules; and conduct an artificial intelligence task by utilizing the optimal proposed model ([0034-0037]).
As to Claim 3:
ZHANG teaches determine the optimal proposed model from the candidate models based on determining a pareto optima between the accuracy rank and the runtime execution rank ([0006-0007], [0039, [0042], and [0059-0061]).
As to Claim 4:
ZHANG teaches utilize a teacher model and a student model during intermediate features distillation conducted for training the candidate modules over a period of time on a distribution of data associated with a dataset ([0065-0066]).
As to Claim 5:
ZHANG teaches utilize a same input to both the teacher model and the student model, wherein the input comprises features of the candidate modules; and utilize an output of the teacher model as a soft label to train the student model ([0006], [0038-0039], and [0065-0066]).
As to Claim 6:
ZHANG teaches identify the candidate modules from the plurality of modules in the module collection of the search space based on a type of a task to be performed ([0034-0036] and [0041-0042]).As to Claim 7:
ZHANG teaches determine whether the plurality of modules in the search space have been updated, new modules have been included in the search space, or a combination thereof ([0040-0041] and [0100]).
As to Claim 8:
ZHANG teaches determine an insertion point within the optimal proposed model for potential substitution with an updated module of the plurality of modules or a new module of the new modules of the search space ([0035-0037] and [0040-0042]).
As to Claim 9:
ZHANG teaches generate a ranking of new candidate modules to substitute a module of the optimal proposed model based on application of a metric to the updated module, the new module, or a combination thereof ([0031-0035]).
As to Claim 10:
ZHANG teaches determine an accuracy rank for the new candidate modules based on training the new candidate modules for a period of time; and execute new candidate models including the new candidate modules on the deep learning accelerator to determine a runtime execution rank for each of the new candidate models ([0037-0039] and [0042]).
As to Claim 11:
ZHANG teaches determine a new optimal proposed model based on the accuracy rank for the new candidate modules and the runtime execution rank for each of the new candidate models ([0034-0037] and [0041-0042]).As to Claim 12:
ZHANG teaches identify layers of the existing artificial intelligence model as candidates for substitution ([0034-0037] and [0041-0042]).As to Claim 13:
ZHANG teaches generate the optimal proposed model from the existing artificial intelligence model ([0071-0074]).
As to Claim 14:
ZHANG teaches a method, comprising:
identifying, by utilizing a neural network, a task to be completed ([0002-0003] and [0031]);
identifying, based on a characteristic of the task, candidate modules of the plurality of modules in a repository for substituting at least a portion of a block within an existing artificial intelligence model ([0035-0037] and [0040-0042]);
training the candidate modules using intermediate features distillation on a distribution of data associated with a dataset ([0031-0032,] [0058-0061], and [0063]);
determining, based on the training, an accuracy rank for each of the candidate modules ([0003], [0037], [0039], [0059], and [0122]);
executing candidate models including the candidate modules on a deep learning accelerator to determine a runtime execution rank for each of the candidate models ([0120-0124]);
determining an optimal proposed model from the candidate models based on the accuracy rank and the runtime execution rank, wherein the optimal proposed model includes a portion of the existing artificial intelligence model and includes a substitution of the portion of the block with a candidate module of the plurality of candidate modules ([0071-0072] and [0099-0102]).
As to Claim 15:
ZHANG teaches executing the task by utilizing the optimal proposed model ([0002-0003] and [0099-0100]).
As to Claim 16:
ZHANG teaches dynamically adjusting the optimal proposed model in real-time as the plurality of modules change ([0040], [0100], and [0122]).
As to Claim 17:
ZHANG teaches training each candidate module using intermediate features distillation without training an entirety of the candidate model including each candidate module ([0034-0037]).
As to Claim 18:
ZHANG teaches identifying the portion of the block of the existing artificial intelligence model to substitute based on a characteristic of a dataset for training the existing artificial intelligence model, a characteristic of the deep learning accelerator, or a combination thereof ([0065-0066] and [0099-0100]).
As to Claim 19:
ZHANG teaches identifying a top-k set of candidate models of the candidate models for execution on the deep learning accelerator ([0037] and 0042).
As to Claim 20:
ZHANG teaches a device (Abstract), comprising:
a memory (Fig. 17); and
a processor (Fig. 17);
wherein the processor is configured to search, by utilizing a neural network, for a plurality of modules in a plurality of repositories ([0002-0003] and [0035]);
wherein the processor is configured to analyze first characteristics of the plurality of modules for inclusion in a search space ([0035-0037], [0039], [0066], and [0074]);
selecting a set of candidate modules of the plurality of modules in the plurality of repositories based on a matching of the first characteristics with second characteristics associated with a task to be completed ([0035-0037]);
wherein the processor is configured to determine an optimal proposed model including at least one candidate module from the set of candidate modules based on an accuracy rank and a runtime execution rank of the candidate modules in the set of candidate modules (([0003], [0037], [0039], [0059], [0071-0072] and [0099-0102]); and
wherein the processor is configured to execute the task using the optimal proposed model ([0120-0124]).
Conclusion
4. The prior art made of record, listed on PTO 892 provided to Applicant is considered to have relevancy to the claimed invention. Applicant should review each identified reference carefully before responding to this office action to properly advance the case in light of the prior art.
Contact information
5. Any inquiry concerning this communication or earlier communications from the
examiner should be directed to MAIKHANH NGUYEN whose telephone number is (571) 272-4093. The examiner can normally be reached on Monday-Friday (8:00 am – 5:30 pm). If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, TAMARA KYLE can be reached at (571)272-4241.
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 or Private PAIR to authorized users only. Should you have questions about access to Patent Center or 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.
/MAIKHANH NGUYEN/Primary Examiner, Art Unit 2144
Prosecution Insights