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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 06/04/2026 has been entered.
Claim Status
Claims 1-20 are currently pending and are presented for examination on the merits.
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 1-20 are rejected under 35 U.S.C. § 101, because they recite non-patentable subject matter under the 2019 PEG, October update. The claimed invention is directed to a judicial exception (e.g., an abstract idea, etc.) without practical application or significantly more.
More particularly, when considering subject matter eligibility under 35 U.S.C. 101, it must be determined whether the claim is directed to one of the four statutory categories of invention, i.e., process, machine, manufacture, or composition of matter. If the claim does fall within one of the statutory categories, it must then be determined whether the claim is directed to a judicial exception (i.e., law of nature, natural phenomenon, and abstract idea), and if so, it must additionally be determined whether the claim is a patent-eligible application of the exception. If an abstract idea is present in the claim, any element or combination of elements in the claim must be sufficient to ensure that the claim amounts to significantly more than the abstract idea itself. Broad categories of abstract ideas include fundamental economic practices, certain methods of organizing human activities, an idea itself, and mathematical relationships/formulas. See, generally Alice Corporation Pty. Ltd. v. CLS Bank International, et al., 573 U.S. __ (2014) (citing Mayo Collaborative Servs. v. Prometheus Labs., Inc.,132 S. Ct. 1289, 1294, 1297-98 (2012)); Federal Register notice titled 2014 Interim Guidance on Patent Subject Matter Eligibility (79 FR 74618), which is found at: http:// www. gpo.gov/fdsys/pkg/FR-2014-12-16/pdf/2014-29414.pdf; 2015 Update to the Interim Guidance; the 2019 Revised Patent Subject Matter Eligibility Guidance, Fed. Reg., Vol. 84, No. 4, January 7, 2019; and associated Office memoranda.
Under the 2019 PEG, step 2a-prong 1, Claims 1-20 recite a judicial exception(s), including a method of organizing human activity (e.g. fundamental economic principle). More particularly, the entirety of the method steps is directed towards, in neural network models, organizing data, identifying valid data, identifying sparsity of data, and rearranging data, so as to reduce computation, effort, and resources. The claims further employ algorithms to generate the model’s output. All of these are abstract ideas in and of themselves. Moreover, these are long standing commercial practices previously performed by humans (e.g., data processors, statisticians, actuaries, etc.) manually and via mental steps. In conventional computing, for example, compacting (or compressing) a hard drive has long included rearranging data on the hard drive to increase space on the drive and improve system performance. As such, the inventions include an abstract idea under the 2019 PEG, and Alice Corporation.
Under step 2a-prong 2, the claims fail to recite a practical application of the exception, because the extraneous limitations (e.g., the structure -a neural network, a non-transitory computer-readable medium, an apparatus, memory, processor, and the methodology including consideration of valid and invalid data, rules for rearranging, and specific placement of rearranged data) merely add insignificant extra-solution activity to the judicial exception (MPEP 2106.05(g), generally link the use of the judicial exception to a particular technological environment or field of use (MPEP 2106.05(h)), and/or generally instruct an artisan to apply it (the method) across generic computing technology. A claim does not cease to be abstract for section 101 purposes simply because the claim confines the abstract idea to a particular technological environment in order to effectuate a real-world benefit. See Alice, 573 U.S. at 222; BSG Tech LLC v. BuySeasons, Inc., 899 F.3d 1281, 1287 (Fed. Cir. 2018); buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1353 (Fed. Cir. 2014). “[I]t is not enough, however, to merely improve a fundamental practice or abstract process by invoking a computer merely as a tool.” Customedia Techs., LLC v. Dish Network Corp., 951 F.3d 1359, 1364 (Fed. Cir. 2020) (citations omitted). More particularly, the claims fail to recite an improvement to the functioning of a computer or technology (under MPEP § 2106.05(a)), the use of a particular machine (under § 2106.05(b)), effect a transformation or reduction of a particular article (§ 2106.05(c)), or apply the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment (§ 2106.05(e)).
Under part 2b, the extraneous limitations offered by the dependent claims either further delineate the abstract idea (express a further step of the algorithm), recite insignificant extra-solution activity, or instruct the artisan to apply it (the abstract idea) across generic computing technology. The claims as a whole, do not amount to significantly more than the abstract idea itself. This is because no one claim effects an improvement to another technology or technical field, an improvement to the functioning of a computer itself, or move beyond a general link of the use of the abstract idea to a particular technological environment. For example, it is not disclose why the specific locations the data are moved to would provide significantly more than the same compression being provided at an alternative location. Viewing the limitations as an ordered combination does not add anything further than looking at the limitations individually. Under Alice, merely applying or executing the abstract idea on one or more generic computer system (e.g., a computer system comprising a generic database; a generic element (NIC) for providing website access, etc.; a generic element for receiving user input; and a generic display on the computer, in any of their forms) to carry out the abstract idea more efficiently fails to cure patent ineligibility. See, e.g., Content Extraction, 776 F.3d at 1347 (claims reciting a “scanner” are nevertheless directed to an abstract idea); Mortg. Grader, Inc. v. First Choice Loan Serv. Inc., 811 F.3d 1314, 1324–25 (Fed. Cir. 2016) (claims reciting an “interface,” “network,” and a “database” are nevertheless directed to an abstract idea).
Courts have recognized the following computer functions to be well‐understood, routine, and conventional functions when they are claimed in a merely generic manner: performing repetitive calculations, receiving, processing, and storing data, electronically scanning or extracting data from a physical document, electronic recordkeeping, automating mental tasks, and receiving or transmitting data over a network, e.g., using the Internet to gather data, MPEP 2106.05(d), wherein the italicized tasks are particularly germane to the instant invention.
Claim Rejections - 35 USC § 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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
a. Determining the scope and contents of the prior art.
b. Ascertaining the differences between the prior art and the claims at issue.
c. Resolving the level of ordinary skill in the pertinent art.
d. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims under pre-AIA 35 U.S.C. 103(a), the examiner presumes that the subject matter of the various claims was commonly owned at the time any inventions covered therein were made absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and invention dates of each claim that was not commonly owned at the time a later invention was made in order for the examiner to consider the applicability of pre-AIA 35 U.S.C. 103(c) and potential pre-AIA 35 U.S.C. 102(e), (f) or (g) prior art under pre-AIA 35 U.S.C. 103(a).
Claims 1-20 are rejected under § 103 as being unpatentable over Kreeger, in view of US 2019/0362235 to Xu et al.
With respect to Claims 1 and 11, Kreeger teaches an apparatus for processing data in a neural network comprising a memory and processor ([0157]) configured to perform a method of processing data (FIG. 10; [0089]) in a neural network, the method comprising: identifying a sparsity among information included in input data ([0081];[0089]), based on valid information or invalid information included in the input data ([0089], “. . . whether there is valid data or zero data in that position”); rearranging (“reorder”, “reconstruct”) the input data ([0089]), based on the sparsity indicating a distribution of the invalid values included in the input data ([0081-82];[0089]) so that the input data are repositioned where a first input data at a first location is rearranged to a second location different from the first location and the rearranging changes a memory access pattern of an operation ([0089]; [0145]; FIGS. 15,16); and generating, by performing an operation on the rearranged input data in the neural network, an output data ([0072];[0089]; [0145]; FIGS. 15,16). The system in Kreeger is configured to perform the recited steps of the invention, however, it does not use the exact same language. Please note that the applied reference(s) need not use the same terminology, or disclose the limitation verbatim. Any differences between Kreeger and the instant invention are obvious, under one or more rationales held in KSR Int’l Co. v. Teleflex Inc., 127 S. Ct. 1727 (2007). For example, whereas Kreeger teaches reordering the input data and performing an action on the reordered data so as to generate output data, it would have been obvious to try “rearranging” instead of “reorder” or “an output data” instead of “an output” amongst a finite set of synonymous or similar word choices. As such, it would have been obvious to modify Kreeger to teach the instant invention.
Kreeger fails to expressly teach, but Xu teaches “so that the input data are repositioned and computation needed to generate output data is reduced while maintaining accuracy of the output data. ([0046]) Xu discusses the need to maintain accuracy in neural networking. [0003] It would have been obvious to one of ordinary skill in the art to modify Kreeger to include this limitation taught by Xu.
With respect to Claims 2 and 12, Kreeger teaches wherein rearranging the input data comprises rearranging rows included in the input data based on a number of invalid values included in each of the rows of the input data. [0077-78];[0089]; FIG. 15
With respect to Claims 3 and 13, Kreeger fails to expressly teach wherein rearranging the input data comprises performing rearrangement such that a first row of the input data comprising the most invalid values among the rows of the input data is adjacent to a second row of the input data comprising the least invalid values among the rows of the input data. However, the specific placement of such rows of rearranged data is a matter of design choice. Alternatively, the adjacent location of the two rows is obvious to try, as there are only a finite number of rows and columns. See, KSR international Co. v. Teleflex Inc. It would have been obvious to one of ordinary skill in the art to modify Kreeger to include a first row comprising the most invalid (e.g., zero) values adjacent a second row comprising the least invalid values.
With respect to Claims 4 and 14, Kreeger teaches wherein rearranging the input data comprises shifting elements of columns included in the input data according to a first rule. [0089];[0077];FIG. 15
With respect to Claims 5 and 15, Kreeger teaches wherein the first rule comprises shifting the elements of the columns included the input data in a same direction by a particular size, and the first rule is periodically applied to the columns included in the input data. [0077]; [0089]; FIGS. 9,15
With respect to Claims 6 and 16, Kreeger teaches wherein rearranging the input data comprises rearranging columns included in the input data to skip processing with respect to at least one column comprising only invalid values. ([0089],“random” teaches skipping a column of data; FIGS. 9, 15).
With respect to Claims 7 and 17, Kreeger fails to expressly teach wherein rearranging the input data comprises shifting a first element of a first column included in the input data to a position corresponding to a last element of a second column of the input data that is adjacent to the first column. However, the specific placement of such rearranged columns of data is a matter of design choice. Alternatively, the adjacent location of the two elements is obvious to try, as there are only a finite number of rows and columns. See, KSR international Co. v. Teleflex Inc. It would have been obvious to one of ordinary skill in the art to modify Kreeger to include a first row comprising the most invalid (e.g., zero) values adjacent a second row comprising the least invalid values.
With respect to Claims 8 and 18, Kreeger teaches wherein generating the output data comprises: applying one or both of a second rule and a third rule to the rearranged input data; and performing a convolution operation on the rearranged input data to which the one or both of the second rule and the third rule is applied and another data. [0081];[0089]
With respect to Claims 9, and 19, Kreeger fails to expressly teach wherein the second rule includes shifting elements of columns, included in the input data, to same positions of an adjacent column, and the third rule comprises shifting elements of columns to transversal positions of an adjacent column. However, the specific placement of such rearranged data is a matter of design choice. Alternatively, the adjacent location of the data is obvious to try, as there are only a finite number of rows and columns. See, KSR international Co. v. Teleflex Inc. It would have been obvious to one of ordinary skill in the art to modify Kreeger to include a second and third rule as recited.
With respect to Claim 10, Kreeger teaches a non-transitory computer-readable recording medium having recorded thereon a program for executing the method of claim 1 on a computer. [0157]
Response to remarks
Applicant’s remarks submitted on 5/4/2026 have been fully considered, but are not persuasive where objections/rejections are maintained. Having considered Applicant’s thorough critique of Kreeger, Examiner maintains the § 103 rejection, because Examiner’s understanding of Kreeger includes a rearranging of input data to remove the zero’s (or invalid data), so as to compress the input data and address “sparsity.” Kreeger teaches wherein at least the zero values are repositioned (i.e., going from their first location to being removed from the input data), which teaches the new limitation added by the instant amendment, based on its broadest reasonable interpretation. Kreeger provides at ¶ [0089] in its entirety as follows:
The embodiments facilitate more efficient data compression. Neural Networks, by their very definition, contain a high degree of sparsity, for the SegNet CNN over 3× the computations involve a zero element. Clearly, having an architecture that can automatically eliminate the excess data movements for zero data, and the redundant multiply by zero for both random and non-random sparsity would result in higher performance and lower power dissipation. Data which is not moved results in a bandwidth reduction and a power savings. Multiplications that do not need to be performed also save power dissipation as well as allowing the multiplier to be utilized for data which is non-zero. The highest bandwidth and computation load in terms of multiply accumulates occurs in the DataStreams exiting the “Reorder” modules in 801 which feed the “Convolve” Modules 802. Automatically compressing the data leaving the reorder module, 801, reduces the bandwidth required to feed the convolve modules as well as reducing the maximum MAC (multiply accumulates) that each convolve performs. There are several possible zero compression schemes that may be performed, what is illustrated is a scheme which takes into account the nature of convolution neural networks. The input to a convolver, 802, consists of a 3-dimensional data structure (Width×Height×Channel). Convolution is defined as multiplying and summing (accumulating) each element of the W×H×C against a Kernel Weight data structure also consisting of (Width×Height×Channel). The data input into the convolver exhibits two types of sparsity—random zeros interspersed in the W×H×C data structure and short “bursts” of zeros across consecutive (W+1)×(H+1)×C data elements. The compressed data structure that is sent from the Reorder Modules to the Convolver modules is detailed in FIG. 9. For every possible 32 values one Bitmask value, 901, is sent followed by any non-zero data values, 902. Each bit position in the bitmask indicates where whether there is valid data or zero data in that position. In the case where there is no zero data, 901 will be all zeros, followed by 32 data values, 902. In the other extreme where there are 32 zero data values, 901 will be all “1” 's and no data values, 902, will follow. In the case there is a mixture of non-zero data values and data values the bitmask, 901, will indicate this and only the non-zero data values will follow in 902. FIG. 10 is the flow chart for the circuitry which resides in 801 the reorder module which performs the compression.
(emphasis added). There, “excess data movements” and multiply of zero references the savings of computational resources. Xu teaches, in a related pruning context, a neural network process that reduces computational resources while maintaining accuracy. Applicant is invited to interview the case to explain the differences between the instant invention and Kreeger. Please note that the applied reference(s) need not use the same terminology, or disclose the limitation verbatim, and also that the entirety of a prior art reference is to be applied to the respective claim(s), such that the pinpoint citations above are exemplary and provided for Applicant’s benefit; other locations within the applied reference(s) may further support the rejection. MPEP 2141.02(VI).
As per § 101, the invention remains directed to compression and rearrangement of input data so as to minimize computational resources. The amendment fails to offer an innovative concept, because compression (e.g., of a hard drive, etc.) has long presented the rearranging of data, so as to reduce computation, space, and efficiency. Applying same in a particular setting/context of neural networking, and more particularly a CNN, would be insufficient where new. However, as the prior art references of record show, this is not novel. (See, Kreeger; see also, US 2016/0358069 to Brothers, and US 10970629 to Dirac (added to the record as a result of an updated search)). The § 101 rejection is maintained as the claims continue to recite an abstract idea without practical application or significantly more. No one claim introduces an innovative concept to effect a practical application by itself.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WILLIAM J JACOB whose telephone number is (571)270-3082. The examiner can normally be reached on M-F 8:00-5:00, alternating Fri. off.
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 Gart can be reached on 5712723955. 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 the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/WILLIAM J JACOB/Examiner, Art Unit 3696
Prosecution Insights