DETAILED ACTION
This office action is responsive to claims 1 - 10 filed in this application Liu et al., U.S. Patent Application No. 18/846,282, (Filed September 12, 2024) claiming priority to PCTCN2021095209 (5/21/2021) claiming priority to CN202011224090.4.
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
The information disclosure statement (IDS) filed on June 7, 2026 is not in compliance with the provisions of 37 CFR 1.97, 1.98 and MPEP § 609 as no copies and translations of the references have been provided. The references listed therein have NOT been considered with the exception of the single US Patent reference, which has been placed in the application file.
Objection to the Drawings
The drawings are objected to under 37 CFR 1.83(a) because they fail to show the reference characters for the elements steps (i.e., S21 – S27) as described in the specification when discussing figure 2. Any structural detail that is essential for a proper understanding of the disclosed invention should be shown in the drawing. MPEP § 608.02(d). Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as "amended." If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either "Replacement Sheet" or "New Sheet" pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Objections
Claim 5 is objected to for the following informality: Claim 5 contains reference characters that correspond to elements recited in the detailed description which appear to refer to the drawings and which are not enclosed within parentheses. See MPEP 608.01(m).
Claim Rejections - 35 USC § 112(b)
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 – 10 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 pre-AIA the applicant regards as the invention.
Claim 1 is rejected as being indefinite. Claim 1 recites “optimizing the intermediate representation file to obtain an optimized intermediate representation file, based on a performance analysis, single-node optimization, and collaborated optimization.” It is unclear if it is the optimizing step, the obtaining step, or both that are based on the “performance analysis, single-node optimization, and collaborated optimization.”
Claims 2 – 7, 9 and 10 are rejected as depending on claim 1. Claims 8 – 10 are rejected for substantially similar reasons.
Claim 1 is rejected as being indefinite. Claim 1 recites “generating a network template file based on hardware interfaces through the optimized intermediate representation file.” It is unclear how the hardware interface on which the generating is based are “through” the representation file.
Claims 2 – 7, 9 and 10 are rejected as depending on claim 1. Claim 8 – 10 are rejected for substantially similar reasons.
Claim 2 is rejected as being indefinite. Claim 2 recites “primary domains” of the abstraction layer, that “the abstraction layer comprises a model, an operator set, fusion blocks, basic layers, and operational operators,” (Five Elements) and then claim 2 recites that each of the Five Elements comprising the abstraction layer have “primary domains.” It is unclear what a “primary domain” is intended to represent. It is also unclear if the “primary domains” of the abstraction layer refer to the “primary domains” listed for each of the Five Elements comprising the abstraction layer or if the abstraction layer has separate, unique, “primary domains” from the “primary domains” of the abstraction layer’s constituent Five Elements.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 is a method claim that states “the method for compiling the neural network according to claim 1, wherein” and then recites various instances of “a description…comprises…”. It is unclear if “a description” is intended to claim a structural description element or a functional step of describing.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 recites “a description of the fusion blocks comprises comprising a block fused from basic layers.” It is unclear how the fusion block description “comprises comprising”.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 recites “a description of the basic layers comprises representing one of the operational operators in the network file.” It is unclear if “description” is being used as a noun or a verb in this limitation. If a noun, it would be unclear how a description can “represent” an operational operator. If a verb, it is unclear if an act of describing that represents operational operators in a network file means writing the operational operators to the network file, creating some data structure on disk/memory to “represent” the operational operator, or if “representing” is intended to have some more particular definition that fails to be described in the specification.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 recites “a description of the operational operators comprises providing a detailed description of the operational operators.” It is unclear what the distinction is between a description and a detailed description.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 recites “a description of the operational operators comprises providing a detailed description of the operational operators.” It is unclear what is being provided, the description or the detailed description, or if the act of providing a detailed description is itself a description.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite for a lack of antecedent basis for the claim term “primary domains” of the model, operator set, fusion blocks, basic layers, and operational operator. No “primary domains” of the model, operator set, fusion blocks, basic layers, and operational operator have been previously introduced in the claim or in a claim from which it depends.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 recites “primary domains of the model comprise a set of fusion blocks, and their intermediate representation.” It is unclear “their” refers to the primary domains, the model, or the set of fusion blocks.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 recites “primary domains of the operator set comprise its version.” It is unclear “its” refers to the primary domains or the operator set.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 recites “a set of layers, and inputs and outputs of the layers.” It is unclear to which layers “the layers” refers, the set of layers, the basic layers, the abstraction layer, or some other layers.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 recites “model parallelisms.” It is unclear what is a “parallelism” is in the context of the claim and specification.
Claims 3 – 7 are rejected as depending on claim 2.
Claim 2 is rejected as being indefinite. Claim 2 recites “model parallelisms.” It is unclear to which model the “parallelism” refers, the previously introduced “the model,” “a complete model,” the “a description of the model,” or “a model.”
Claims 3 – 7 are rejected as depending on claim 2. Claim 4 is rejected for substantially similar reasoning.
Claim 3 is rejected as being indefinite. Claim 3 recites “portraying the performance.” It is unclear whether the “portraying” of the performance means performing, demonstrating, displaying, recording, storing, analysis, permutating, or some other meaning.
Claims 4 – 7 are rejected as depending on claim 3.
Claim 3 is rejected as being indefinite. Claim 3 recites “portraying the performance through performance tests.” It is unclear whether “portraying” is intended to claim depicting, describing, recording, demonstrating, simulating, or performing the performance tests for an observer.
Claims 4 – 7 are rejected as depending on claim 3.
Claim 3 is rejected as being indefinite. Claim 3 recites “generating..performances.” It is unclear whether the performances are simply generated or if the performances are performed.
Claims 4 – 7 are rejected as depending on claim 3.
Claim 3 is rejected as being indefinite. Claim 3 recites “obtaining influence parameters affecting the performance of the operational operators.” It is unclear whether the obtaining affects the performance or whether the influence parameters affect the performance
Claims 4 – 7 are rejected as depending on claim 3.
Claim 3 is rejected as being indefinite. Claim 3 recites “constructing a mathematical model by the influence parameters.” It is unclear how a parameter can construct a mathematical model.
Claims 4 – 7 are rejected as depending on claim 3.
Claim 3 is rejected as being indefinite. Claim 3 recites “constructing a mathematical model by the influence parameters to portray the performance of the operational operators.” It is unclear if the constructing portrays performance or if a mathematical model portrays performance.
Claims 4 – 7 are rejected as depending on claim 3.
Claim 3 is rejected as being indefinite. Claim 3 recites “portray the performance.” It is unclear whether “portray” means performing, demonstrating, displaying, recording, storing, analysis, permutating, or some other meaning.
Claims 4 – 7 are rejected as depending on claim 3.
Claim 4 is rejected as being indefinite. Claim 4 twice recites “portraying.” It is unclear whether “portray” means performing, demonstrating, displaying, recording, storing, analysis, permutating, or some other meaning.
Claim 4 is rejected as being indefinite for a lack of antecedent basis for the claim term “the model parallelisms and operator fusion”. No “model parallelisms and operator fusion” have been previously introduced in the claim or in a claim from which it depends.
Claim 4 is rejected as being indefinite. The term “optimal” in claim 6 is a relative term which renders the claim indefinite. The term "optimal" 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.
Claim 5 is rejected as being indefinite. It is unclear if there is a limit on the computation threshold and/or the umber of basic layers such that the claim proceeds indefinitely or if it has a finite (if unclaimed) number of iterations.
Claim 5 is rejected as being indefinite. Claim 5 recites various logical states without specifying what condition the state results from, such as “if yes,” and “if no.” It is unclear to which condition functions that follow these states refer and thus the metes and bounds of the claim cannot be determined.
Claim 6 is rejected as being indefinite. Claim 6 recites “hiding redundant operations and exposing nodes to be optimized, by the abstraction layer.” It is unclear if the abstraction layer performs the functions of “hiding,” “exposing,” and optimizing, or some subset thereof.
35 USC § 112(f)
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are: “module configured to”, in claim 8.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections 35 U.S.C. §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.
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.
Claim 1 – 6 and 8 – 10 are rejected under 35 U.S.C. 103 as being unpatentable over Venkataramani et al., United States Patent Application Publication No. 2018/0136912 (Published May 17, 2018, filed November 17, 2017) in view of Sui et al., China Patent No. CN110766147A (Published February 7, 2020) and indicated as Foreign Document Number 2 “Xilinx” on 9/12/2024 IDS (“Venkataramani”) (“Sui”).
Claims 1, 9, and 10
With respect to claims 1, 9, and 10 Venkataramani teaches the invention as claimed including a method for compiling a neural network, comprising:
translating a network file into an intermediate representation file;… generating a network template file based on hardware interfaces through the optimized intermediate representation file; compiling the network template file into an executable inference application. {A network file for a deep learning network may be translated into one or more intermediate representations in the form of dataflow and/or control flow graphs with nodes that correspond to the layers of the deep learning network, optimizations may be performed on the individual nodes and on cross-layers to generate code as a network template that is target platform neutral, and then the code is compiled for a particular target platform. Venkataramani at Abstract; id. at ¶¶ 0040 - 0042, 0046, 0048 – 0057.}
However, Venkataramani doesn’t explicitly teach the limitation:
optimizing the intermediate representation file to obtain an optimized intermediate representation file, based on a performance analysis, single-node optimization, and collaborated optimization; {Sui does teach this limitation. Sui teaches that translating a network file to an intermediate representation, as taught in Venkataramani, may include where a graph built from the translation is optimized into a second intermediate template using analysis of and optimizations for individual and multiple nodes in the intermediate graph such as node fusing, decomposing, and layer fusion where the final graph is then compiled for execution on a particular hardware target. Sui at fig. 7, pgs. 42 & 43.
Venkataramani and Sui are analogous art because they are from the “same field of endeavor” and are both from the same “problem-solving area.” Specifically, they are both from the field of neural networks, and both are trying to solve the problem of how to optimize a neural network.
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to combine translating a network file to an intermediate representation, as taught in Venkataramani with performing graph optimizations, as taught in Sui. Venkataramani teaches that different target platforms require different neural network optimizations. Id. at ¶ 0039. Therefore, one having ordinary skill in the art would have been motivated to combine translating a network file to an intermediate representation, as taught in Venkataramani with performing graph optimizations, as taught in Sui, for the purpose of customizing a neural network for operation on a particular target platform.}
Claim 2
With respect to claim 2 Venkataramani and Sui teaches the invention as claimed including:
wherein the network file comprises a network structure and network parameters; the intermediate representation file comprises an abstraction layer, descriptions of the abstraction layer, and primary domains of the abstraction layer; {A network file for a deep learning network may be translated into one or more intermediate representations in the form of dataflow and/or control flow graphs with nodes that correspond to the layers of the deep learning network, optimizations may be performed on the individual nodes and on cross-layers to generate code as a network template that is target platform neutral, and then the code is compiled for a particular target platform. Venkataramani at Abstract; id. at ¶¶ 0040 - 0042, 0046, 0048 – 0057.}
the abstraction layer comprises a model, an operator set, fusion blocks, basic layers, and operational operators; {A graph built from the translation is optimized into a second intermediate template using analysis of and optimizations for individual and multiple nodes in the intermediate graph such as node fusing, decomposing, and layer fusion where the final graph is then compiled for execution on a particular hardware target. Sui at fig. 7, pgs. 42 & 43.}
a description of the model comprises describing a complete model execution flow; a description of the operator set comprises specifying an operator set version; a description of the fusion blocks comprises comprising a block fused from basic layers; a description of the basic layers comprises representing one of the operational operators in the network file; a description of the operational operators comprises providing a detailed description of the operational operators; primary domains of the model comprise a set of fusion blocks, and their intermediate representation; primary domains of the operator set comprise its version and a list of included operators; primary domains of the fusion blocks comprise a set of layers, and inputs and outputs of the layers; primary domains of the basic layers comprise operational operators, inputs, outputs, and model parallelisms; primary domains of the operational operator comprise operator types and operator attributes. {EN: The limitations of claim 2 being by stating “compiling the neural network according to claim 1, wherein” and then describe various contents of the network file, intermediate representation file, and abstraction layer, as well as of descriptions and “primary domains” of the contents. However, claim 2 does not further claim any functional steps associated with these described contents and instead merely describes them as structural elements. Since no further steps of the method claim are required to be performed in claim 2, as well as since none of the structure disclosed in claim 2 modifies the steps in claim 1, the wherein clause of claim 2 fails to give any “meaning and purpose to the manipulative steps” and is interpreted as not limiting the scope of the claims. See MPEP 211104 (I) citing to Griffin v. Bertina, 285 F.3d 1029, 1034, 62 USPQ2d 1431 (Fed. Cir. 2002).
A graph built from the translation is optimized into a second intermediate template using analysis of and optimizations for individual and multiple nodes in the intermediate graph such as node fusing, decomposing, and layer fusion where the final graph is then compiled for execution on a particular hardware target. Sui at fig. 7, pgs. 42 & 43.}
Claim 3
With respect to claim 3 Venkataramani and Sui teaches the invention as claimed including:
wherein the optimizing of the intermediate representation file based on the performance analysis comprises: portraying the performance of the operational operators through performance tests, generating a series of measured performances with varying parameters, obtaining influence parameters affecting the performance of the operational operators, and constructing a mathematical model by the influence parameters to portray the performance of the operational operators. {A graph built from the translation is optimized into a second intermediate template using analysis of and optimizations for individual and multiple nodes in the intermediate graph such as node fusing, decomposing, and layer fusion where the final graph is then compiled for execution on a particular hardware target. Sui at fig. 7, pgs. 42 & 43.}
Claim 4
With respect to claim 4 Venkataramani and Sui teaches the invention as claimed including:
wherein the optimizing of the intermediate representation file based on the single-node optimization comprises: portraying the model parallelisms and operator fusion, selecting an optimal model parallelism for the operational operators, and portraying dimensions of fusion blocks, redundant computational amounts, and performance variation. {A graph built from the translation is optimized into a second intermediate template using analysis of and optimizations for individual and multiple nodes in the intermediate graph such as node fusing, decomposing, and layer fusion where the final graph is then compiled for execution on a particular hardware target. Sui at fig. 7, pgs. 42 & 43.}
Claim 5
With respect to claim 5 Venkataramani and Sui teaches the invention as claimed including:
wherein the optimizing of the intermediate representation file based on the collaborated optimization comprises: S21: reading a next basic layer; S22: determining whether this next basic layer is capable of being fused with a current fusion block; if capable, then performing S23: determining whether this next basic layer is a fully connected layer or a convolutional layer of the neural network; if yes, performing S24: counting a computational amount of this next basic layer and adding it to a current total computational amount, and performing S25: adding this next basic layer to the current fusion block, and proceeding to S27; if no, directly performing S25: adding this next basic layer to the current fusion block and proceeding to S27; if not capable, performing S26: opening a new fusion block; S27: determining whether the current total computational amount of fusion blocks exceeds a computation threshold, if yes, proceeding to S26; if no, returning to S21. {EN: Under the broadest reasonable interpretation there is only one basic layer, it is not capable, a new fusion block is opened and the claim end.
A graph built from the translation is optimized into a second intermediate template using analysis of and optimizations for individual and multiple nodes in the intermediate graph such as node fusing, decomposing, and layer fusion where the final graph is then compiled for execution on a particular hardware target. Sui at fig. 7, pgs. 42 & 43; Venkataramani at ¶ 0023 (threshold for operations).}
Claim 6
With respect to claim 6 Venkataramani and Sui teaches the invention as claimed including:
wherein the generating of the network template file further comprises hiding redundant operations and exposing nodes to be optimized, by the abstraction layer. {A graph built from the translation is optimized into a second intermediate template using analysis of and optimizations for individual and multiple nodes in the intermediate graph such as node fusing, decomposing, and layer fusion where the final graph is then compiled for execution on a particular hardware target. Sui at fig. 7, pgs. 42 & 43.}
Claim 8
With respect to claim 8 Venkataramani teaches the invention as claimed including a system for compiling a neural network, comprising:
a translation module configured to translate a network file into an intermediate representation file; … a file generation module configured to generate a network template file based on hardware interfaces through the optimized intermediate representation file; and a compilation module configured to compile the network template file into an executable inference application. {A network file for a deep learning network may be translated into one or more intermediate representations in the form of dataflow and/or control flow graphs with nodes that correspond to the layers of the deep learning network, optimizations may be performed on the individual nodes and on cross-layers to generate code as a network template that is target platform neutral, and then the code is compiled for a particular target platform. Venkataramani at Abstract; id. at ¶¶ 0040 - 0042, 0046, 0048 – 0057.}
However, Venkataramani doesn’t explicitly teach the limitation:
an optimization module configured to optimize the intermediate representation file to obtain an optimized intermediate representation file, based on a performance analysis, single-node optimization, and collaborated optimization;{Sui does teach this limitation. Sui teaches that translating a network file to an intermediate representation, as taught in Venkataramani, may include where a graph built from the translation is optimized into a second intermediate template using analysis of and optimizations for individual and multiple nodes in the intermediate graph such as node fusing, decomposing, and layer fusion where the final graph is then compiled for execution on a particular hardware target. Sui at fig. 7, pgs. 42 & 43.
Venkataramani and Sui are analogous art because they are from the “same field of endeavor” and are both from the same “problem-solving area.” Specifically, they are both from the field of neural networks, and both are trying to solve the problem of how to optimize a neural network.
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to combine translating a network file to an intermediate representation, as taught in Venkataramani with performing graph optimizations, as taught in Sui. Venkataramani teaches that different target platforms require different neural network optimizations. Id. at ¶ 0039. Therefore, one having ordinary skill in the art would have been motivated to combine translating a network file to an intermediate representation, as taught in Venkataramani with performing graph optimizations, as taught in Sui, for the purpose of customizing a neural network for operation on a particular target platform.}
Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over Venkataramani in view of Sui and Dal et al., United States Patent Application Publication No. 2022/0019463 (Published January 20, 2022, filed May 18, 2021) claiming priority to PCT filed June 18, 2020 (“Dal”)
Claim 7
With respect to claim 7 Venkataramani and Sui teaches the invention as claimed including, with the exception of the limitation:
wherein the network template file is compiled into the executable inference application by a G++ compiler. {Dal does teach this limitation. Dal teaches that compiling a neural network for execution, as taught in Venkataramani and Sui, may include where a g++ compiler is used. Dal at ¶¶ 0039, 0085, 0086, and 0103.
Venkataramani, Sui, and Dal are analogous art because they are from the “same field of endeavor” and are both from the same “problem-solving area.” Specifically, they are both from the field of neural networks, and both are trying to solve the problem of how to comile a neural network.
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to combine compiling a neural network for execution, as taught in Venkataramani and Sui with using a a g++ compile, as taught in Dal. Dal teaches that in certain circumstances a g++ compiler may produce superior results to other compilers. Id. at ¶ 0103. Therefore, one having ordinary skill in the art would have been motivated to combine compiling a neural network for execution, as taught in Venkataramani and Sui with using a a g++ compile, as taught in Dal, for the purpose of implementing an optimal compilation for the neural network.}
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to THEODORE E HEBERT whose telephone number is (571)270-1409. The examiner can normally be reached on Monday to Friday 9:00 a.m. to 6:00 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, Lewis Bullock can be reached on 571-272-3759. 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.
//T.H./ June 27, 2026
Examiner, Art Unit 2199
/LEWIS A BULLOCK JR/Supervisory Patent Examiner, Art Unit 2199
Prosecution Insights