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 and the preliminary amendment filed 12/15/2023.
Claims 26-50 are presented for examination. Claims 1-25 have been cancelled. Claims 26 and 38 are independent Claims.
Drawings
2. The drawings filed 12/15/2023 are acceptable for examination purposes.
Information Disclosure Statement
3. The Applicant’s Information Disclosure Statement filed 12/15/2023 has been received, entered into the record, and considered.
Claim Rejections - 35 USC § 101
4. 35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 26-36 and 50 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
Regarding independent claim 26, the claim recites a “hardware-aware cost model”. However, as currently recited the “hardware-aware cost model” comprises only computer software components. Thus, the claim is software per se and does not fall within any of the four enumerated categories of patentable subject matter in section 101.
Accordingly, claim 26 fails to recite statutory subject matter under 35 U.S.C. 101.
For the same reasons discussed supra with respect to independent claim 26, claims 27-36 fall outside the scope of § 101.
Regarding dependent claim 50, the claim recites a “computer-readable storage medium.” The broadest reasonable interpretation of a claim drawn to a computer-readable storage medium typically covers forms of non-transitory tangible media and transitory propagating signals per se in view of the ordinary and customary meaning of computer readable storage medium. See Ex parte Mehwerter, App. No. 2012-007692 (PTAB 2013)(precedential).
The Specification fails to provide a definition of computer-readable storage medium that excludes transitory propagating signals. Thus, the recited “computer-readable storage medium” is interpreted to include nonstatutory subject matter (e.g., signals, carrier waves, etc.).
The Examiner suggests amending the above claim to explicitly exclude signals (e.g., by adding the phrase “non-transitory”) to obviate the rejection.
To expedite a complete examination of the instant application, the claims rejected under 35 U.S.C. §101 above are further rejected as set forth below in anticipation of Applicant amending these claims to place them within the four statutory categories of invention.
Claim Rejections - 35 USC § 103
5. In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent may not be obtained though the invention is not identically disclosed or described as set forth in section 102 of this title, if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains. Patentability shall not be negatived by the manner in which the invention was made.
Claims 26-34, 36, 38-46, 48, and 50 are rejected under 35 U.S.C. 103 as being unpatentable over Narayan et al. (US 20230037227) in view of Ebrahimi Afrouzi et al. (US 20220066456).
It is noted that any citations to specific, pages, columns, paragraphs, lines, or figures in the prior art references and any interpretation of the reference should not be considered to be limiting in any way. A reference is relevant for all it contains and may be relied upon for all that it would have reasonably suggested to one having ordinary skill in the art. See MPEP 2123.
As to claim 26:
Narayan teaches a hardware-aware cost model for optimizing inference of a deep neural network ([0066]) comprising:
a computation cost estimator configured to compute estimated computation cost based on input tensor, weight tensor and output tensor from the DNN ([0042], [0045], and [0121-0125]); and
a memory/cache cost estimator configured to perform memory/cache cost estimation strategy based on hardware specifications ([0057], [0066], and [0078—0080]),
wherein the hardware-aware cost model is used to perform performance simulation on target hardware to quantization as required for converting a conventional precision inference model to an optimized inference model based on the result of the performance simulation ([0042-0043], and [0063-0066]).
Narayan, however, does not explicitly teach, Ebrahimi Afrouzi teaches the use of dynamic quantization knobs ([0286] and [0291]).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to modify Narayan with Ebrahimi Afrouzi because it would have provided the enhanced capability for reducing the error of the network.
As to claim 27:
Narayan teaches the quantization is post-training quantization ([0043] and [0049-0050]).
As to claim 28:
Narayan teaches the conventional precision inference model comprises FP32 model ([0132]).
As to claim 29:
Narayan teaches the optimized inference model comprises Bfloat16 model, Bfloat8 model and INT8 model ([0109], [0130], and [0154]).
As to claim 30:
Narayan teaches the hardware-aware cost model is constructed on top of intermediate representation (IR) builder ([0109], [0130], and [0154]).As to claim 31:
Narayan teaches the input tensor has four dimensions and is represented as input (N, C.sub.in, H.sub.in, W.sub.in), wherein N is batch size, C.sub.in is input channel count, H.sub.in is height of input data and W.sub.in is width of input data ([0039] and [0045]).
As to claim 32:
Narayan teaches the weight tensor has four dimensions and is represented as input (C.sub.out, C.sub.in, KH, KW), wherein C.sub.out is output channel count, C.sub.in is input channel count, KH is kernel height and KW is kernel width ([0045-0046], [0052, and [0134]).
As to claim 33:
Narayan teaches the output tensor has four dimensions and is represented as input (N, C.sub.out, H.sub.out, W.sub.out), wherein N is batch size, C.sub.out is output channel count, H.sub.out is height of output data and W.sub.out is width of output data ([0045-] and [0078]).
As to claim 34:
Narayan teaches the computation cost estimator is configured to compute the estimated computation cost T by using the following equation: T=2×N×Cout×Hout×Wout×Cin×KH×KW÷(strideofconvolution). ([0063] and [0066]).As to claim 36:
Narayan teaches the memory/cache cost estimator is configured to perform the memory/cache cost estimation strategy comprising: reading weight tensor from a cache or a memory; and caching the weight tensor until there is no free space in the cache ([0125] and [0184]).
As to claims 38-46 and 48:
Refer to the discussion of claims 26-34 and 36 above, respectively, for rejections. Claims 38-46 and 48 are the same as claims 26-34 and 36, except claims 38-46 and 48 are “method” claims and claims 26-34 and 36 are “hardware-aware cost model” claims.
As to claim 50:
Narayan teaches A computer-readable storage medium with program instructions stored thereon which, when executed by a processor, cause the processor to implement the method of claim 38 ([0025] and [0049]).
Allowable Subject Matter
6. Claims 35 and 37are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims, subject to the 101 rejections detailed above, subject to the results of a final search by the Examiner.
Claims 47 and 49 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims, subject to the results of a final search by the Examiner.
Conclusion
7. 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
8. 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