DETAILED ACTION
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 .
The following action is in response to the original filing of 09/26/2023.
Claims 1-20 are pending and have been considered below.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Regarding claims 1, 10 and 11, each of independent claims 1, 10 and 11 recite the phrase “approximates sequential operations” and “approximate floating point precision” (claim 1: line 7, 9-10; claim 10: lines 6, 9; claim 11: lines 5, 8-9). The term “approximate” in both cases+ is not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. This term fails to provide a quantitative metric or threshold to determine when the conditions of the limitations that follow qualify for operation.
Regarding claims 1, 10 and 11, claim 1 recites the limitations: “receive data” (line 3), “convert the data” (line 5), “conversion of the received data” (lines 7-8), “enables the data” (line 9), and “process the data” (line 11); claim 10 similarly recites: “receiving data” (line 3), “converting the received data” (line 4) and “enables the data” (line 8); claim 11 similarly recites: “receiving data” (line 2), “converting the received data” (line 3), “processing the data” (line 8). Applicant’s use of “data” and “received data” render each claim unclear because it is unclear as to which state of data is being discussed. For example, the data of claim 1 is initially received (in a first fixed point format), intermediately converted (approximating a floating point format) and ultimately converted (in a second fixed point format). By using the terms “received data” and “the data” it causes the claim to be unclear as to which particular state of data the limitations refer.
Regarding claims 2-4, 9, 13-15 and 19, each of claims 2-4, 9, 13-15 and 19 recite “data” or “the data” and are rejected for similar reasons as presented with regards to claims 1, 10 and 11.
Regarding claims 8 and 19, each of claims 8 and 19 recite the phrase “a high precision parameter that approximates floating point precision”. The terms “high precision” and “approximates” are not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. This term fails to provide a quantitative metric or threshold to determine when the conditions of the limitations that follow qualify for operation.
Regarding claims 2-9 and 12-20, claims 2-9, which depend from claim 1, and claims 12-20, which depend from claim 11, are similarly rejected for not correcting the corresponding issues of their respective parent claims.
Claim Rejections - 35 USC § 101
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 1-9 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
Regarding claim 1, claim 1 recites: “A computing system, comprising: a hardware accelerator configured to: ..”. The claim does not fall within at least one of the four categories of patent eligible subject matter because while the specification states that a hardware accelerator may include hardware (pp. 1, pp. 16: “Hardware accelerators may include, for example, a neural processing unit (NPU) and/or a graphics processing unit (GPU).”), the specification only provides examples and is non-limiting. Ab broadest reasonable interpretation in light of the specification of a hardware accelerator yields an interpretation that includes software per se. Accordingly, the computing system of claim 1 is comprised of software per se, a non-statutory subject matter.
Regarding claims 2-9, claims 2-9 depend from claim 1 and are similarly rejected.
Claims 10-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to abstract ideas without significantly more.
Regarding claims 10 and 11:
Step 1, MPEP 2106.03:
These limitations have been determined, under Step 1, to be statutory categories of invention:
A computer-readable storage medium having instructions recorded thereon [..] (claim 10)
A method: [..] (claim 11)
Step 2A Prong One MPEP 2106.04, 2106.04(a):
These limitations represent, under Step 2A Prong One, mental processes such as concepts that can be practically performed in the human mind, or by a human using pen and paper as a physical aid, including observations, evaluations, judgments and opinions, MPEP 2106.04(a)(2)(III), for example a person can, with pen and paper, convert data from a first fixed point format to a second fixed point format according to a set of operations with adequate precision to approximate floating point precision:
[..] converting the received data from the first fixed point format to a second fixed point format in a first operation with a first set of parameters; [..]
[..] wherein the first operation approximates sequential operations comprising an intermediate conversion of the first fixed point format to a floating point format and conversion of the floating point format to the second fixed point format [..]
[..] which enables the data in the second fixed point format to approximate floating point precision [..] (claim 10)
[..] processing the data in the second fixed point format to approximate floating point precision [..] (claim 11)
Step 2A Prong Two, MPEP 2106.04(d):
These limitations represent, under Step 2A Prong Two, mere instructions to implement the abstract idea using generic computing tools, MPEP 2106.05(f):
[..] when executed by a hardware accelerator, implements a method [..] (claim 10)
These limitations represent, under Step 2A Prong Two, mere data gathering, MPEP 2106.05:
[..] receiving data in a first fixed point format [..]
Step 2B, MPEP 2106.05:
These limitations are considered, under Step 2B, insignificant extra-solution activity as being recited at a high level of generality, MPEP 2106.05(d):
[..] when executed by a hardware accelerator, implements a method [..] (claim 10)
These limitations are considered, under Step 2B, insignificant extra-solution activity of data gathering/selecting a particular type of data, MPEP 2106.05(g):
[..] receiving data in a first fixed point format [..]
Regarding dependent claim 12, this dependent claim additionally recites that the first operation is implemented by a hardware accelerator. The analysis incorporates the Step analysis of its respective parent. The additional limitations represent, under Step 2A Prong Two, mere instructions to implement the abstract idea using generic computing tools, MPEP 2106.05(f); under Step 2B, insignificant extra-solution activity as being recited at a high level of generality, MPEP 2106.05(d).
Regarding dependent claims 13-15, this dependent claim further recites limitations that detect a type/format of data, generate an operator to configure the conversion (claim 13) and that the first operation comprises quantization/dequantization and the approximated sequential operations comprise quantization and pre/post processing of the data (claims 14 and 15). The analysis incorporates the Step analysis of its respective parent. These limitations represent mental processes such as concepts that can be practically performed in the human mind, or by a human using pen and paper as a physical aid, including observations, evaluations, judgments and opinions, MPEP 2106.04(a)(2)(III). Under Steps 2A Prong Two and Step 2B, all respective limitations are part of the abstract idea.
Regarding dependent claims 16 and 20, these dependent claims additionally recite limitations that defining what the data of the first set of parameters comprise (claim 16) and what the data input comprises (claim 20). The analysis incorporates the Step analysis of its respective parent. The additional limitations represent, under Step 2A Prong Two, mere data gathering, MPEP 2106.05; under Step 2B, insignificant extra-solution activity of data gathering/selecting a particular type of data, MPEP 2106.05(g).
Regarding dependent claim 17, this dependent claim additionally recites that the first operation comprises multiplication and addition. The analysis incorporates the Step analysis of its respective parent. These limitations represent under Step 2A Prong One, mathematical concepts such as mathematical relationships, mathematical formulas or equations, or mathematical calculations, MPEP 2106.04(a)(2)(I). Under Steps 2A Prong Two and Step 2B, all respective limitations are part of the abstract idea.
Regarding dependent claims 18 and 19, these dependent claims additionally recite limitations that the first fixed point format is unsigned integer and the second fixed point format is integer (claim 18) and that the at least one parameter is a high precision parameter approximating floating point (claim 19. The analysis incorporates the Step analysis of its respective parent. The additional limitations represent, under Step 2A Prong Two, mere instructions to implement the abstract idea using generic computing tools, MPEP 2106.05(f); under Step 2B, mere instructions to apply to obtain a solution/outcome, MPEP 2106.05(f).
Claim Rejections - 35 USC § 102
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 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 1-2, 4-13 and 15-20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Michigami et al., US 2025/0181315 A1 [MICHIGAMI].
Regarding claim 1, MICHIGAMI discloses a system, comprising:
a hardware accelerator (Fig. 1 10 neural network circuit) configured to:
receive data in a first fixed point format different from a second fixed point format that a machine learning (ML) model is configured to process (Fig. 1 18->20, Fig. 2 18->20, pp. 53-55: receive int8/uint8 signed and unsigned fixed point number, pp. 105: a conversion unit that converts a signed fixed-point number and an unsigned fixed-point number into an extended signed fixed-point number);
convert the data from the first fixed point format to the second fixed point format in a first operation with a first set of parameters (pp. 65: intNi signed fixed point number, unitNu unsigned fixed-point number, pp. 75: intNi and uintNu converted into extended signed fixed-point number intN), wherein the first operation approximates sequential operations comprising an intermediate conversion of the received data to a floating point format and conversion of the floating point format to the second fixed point format (pp. 75: both operations performed as single in subsequent operations, broadly approximating sequential operations via intermediate conversion), which enables the data in the second fixed point format to approximate floating point precision (pp. 75: “possible to prevent a decrease in accuracy”, broadly enabled to approximate floating point precision); and
implement the ML model to process the data in the second fixed point format (Fig. 1, pp. 29: neural network circuit using the converted data).
Regarding claim 2, MICHIGAMI discloses the computing system of claim 1, wherein the hardware accelerator is further configured to:
detect at least one of a type of the data or a format of the data (pp. 55: signed fixed-point and unsigned-fixed point);
generate, based at least on the detection, an operation descriptor indicating the first operation and the first set of parameters (pp. 55: determine absence of sign of input 8-bit wide fixed-point number on basis of a control signal); and
provide the operation descriptor to configure the conversion (pp. 55: perform conversion).
Regarding claim 4, MICHIGAMI discloses the computing system of claim 1, wherein the first operation comprises modified quantization (pp. 45: quantization operation) and the approximated sequential operations comprise preprocessing of data for the ML model and quantization (pp. 35-36: data conversion preprocessing).
Regarding claim 5, MICHIGAMI discloses the computing system of claim 4, wherein the first set of parameters for the modified quantization comprises at least one parameter for the quantization (pp. 45: operators) and at least one parameter for the preprocessing (pp. 75: including all numerical bits in preprocessing).
Regarding claim 6, MICHIGAMI discloses the computing system of claim 1, wherein the first operation comprises multiplication and addition (pp. 8: a product-sum operator).
Regarding claim 7, MICHIGAMI discloses the computing system of claim 1, wherein the first fixed point format comprises unsigned integer format and the second fixed point format comprises integer format (pp. 65).
Regarding claim 8, MICHIGAMI discloses the computing system of claim 1, wherein at least one parameter in the first set of parameters is a high precision parameter that approximates floating point precision (pp. 75: including all numerical bits reduces accuracy loss, broadly approximating high precision floating point).
Regarding claim 9, MICHIGAMI discloses the computing system of claim 1, wherein the data comprises raw image data (pp. 29).
Regarding claims 10 and 11, claims 10 and 11 recite limitations similar to claim 1 and are similarly rejected.
Regarding claim 12, claim 12 recites limitations similar to claim 1 and is similarly rejected.
Regarding claim 13, claim 13 recites limitations similar to claim 2 and is similarly rejected.
Regarding claim 15, claim 15 recites limitations similar to claim 4 and is similarly rejected.
Regarding claim 16, claim 16 recites limitations similar to claim 5 and is similarly rejected.
Regarding claim 17, claim 17 recites limitations similar to claim 6 and is similarly rejected.
Regarding claim 18, claim 18 recites limitations similar to claim 7 and is similarly rejected.
Regarding claim 19, claim 19 recites limitations similar to claim 8 and is similarly rejected.
Regarding claim 20, claim 20 recites limitations similar to claim 9 and is similarly rejected.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 3 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over MICHIGAMI in view of Papadopoulos et al., US 2025/0006232 A1 [“PAPDOPOULOS”].
Regarding claim 3, MICHIGAMI discloses the computing system of claim 1, wherein the first operation comprises the approximated sequential operations comprise quantization and postprocessing of data for the ML model (pp. 45: quantization postprocessing).
MICHIGAMI fails to disclose wherein the first operation comprises modified dequantization.
PAPADOPOULOS discloses methods for hardware accelerator improvements involving data conversion (pp. 13-14), an analogous art. In particular, PAPADOPOULOS discloses performing a modified dequantization operation following quantization and postprocessing of data (pp. 26: performing a quantization operation, pp. 56: performing dequantization operation). Therefore it would have been obvious to one having ordinary skill in the art and the teachings of MICHIGAMI and PAPADOPOULOS before them before the effective filing of the claimed invention to combine the operation comprising modified dequantization, as taught by PAPADOPOULOS, with MICHIGAMI’s operation comprising quantization and postprocessing of data. One would have been motivated to do this to provide higher precision data for use by the machine learning model, as suggested by PAPADOPOULOS (pp. 56).
Regarding claim 14, claim 14 recites limitations similar to claim 3 and is similarly rejected.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
David; Nir et al.
US 20250103287 A1
DIRECT FIXED POINT TO FIXED POINT DATA CONVERSION APPROXIMATING FLOATING POINT PRECISION IN HARDWARE ACCELERATOR
Zhang; Wenhao et al.
US 20230410255 A1
DECREASED QUANTIZATION LATENCY
Göddeke, Dominik, Robert Strzodka, and Stefan Turek. "Performance and accuracy of hardware-oriented native-, emulated-and mixed-precision solvers in FEM simulations." International Journal of Parallel, Emergent and Distributed Systems 22.4 (2007): 221-256.
Kneusel, Ronald T. "Fixed-Point Numbers." Numbers and Computers. Cham: Springer International Publishing, 2017. 183-213.
Badaroux, Marie, and Frédéric Pétrot. "Arbitrary and Variable Precision Floating-Point Arithmetic Support in Dynamic Binary Translation." Proceedings of the 26th Asia and South Pacific Design Automation Conference. 2021.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ANDREW L TANK whose telephone number is (571)270-1692. The examiner can normally be reached Monday-Thursday 9a-6p.
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, Matthew Ell can be reached at 571-270-3264. 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.
/ANDREW L TANK/ Primary Examiner, Art Unit 2141
Prosecution Insights