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 . This Office Action is in reply to communication filed on 10/16/2024. Claimed priority is granted from Foreign Application Number KR10-2023-0138858 filed on 10/17/2023.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 10/16/2024 was filed simultaneously with the application. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
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, 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.
Claims 1, 4, 8-10, 14, 16, 17, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Wang, US 2021/0149634 A1, in view of Mangnall, US 2022/022987 A1, and further in view of Bar-On, US 2023/0229732 A1.
Regarding claim 1, the combination Wang, Mangnall, and Bar-On teaches:
A method of approximating comprising a nonlinear function in units of a plurality of segments (Bar-On, 0022-0034, 0039-0042; Figs. 1-2: input intervals and respective linear approximations), the method comprising:
detecting an input included in a shortest segment among the plurality of segments (Bar-On 0028, 0068: expressly identifies a smallest existing segment during construction. Selecting an input within a shortest final segment is the proposed modification discussed below);
shifting the nonlinear function such that the input is zero in a floating point format (Wang, 0035-0041, Fig. 2: translates a critical input region toward zero; and 0090 and Fig. 8: Expressly permits subtraction);
defining the plurality of segments based on at least one bit of the floating point format (Wang, 0026-0028, 0091-0092, Figs. 1A, 8: translated-input bits address tables. Mangnall, 0041, 0051l Figs. 3, 6: exponent/mantissa-based selection); and
approximating the plurality of segments into a plurality of linear functions, respectively (Wang, 0032; Fig. 1B, and Mangnall, 0043: coefficient-based linear evaluation and interpolation).
Accordingly, a person of ordinary skill would have been motivated to combine Wang with Mangnall to improve approximation accuracy while controlling lookup-table storage. Wang translates a selected input region into a region having greater floating-point resolution (0026, 0040-0041; Fig. 2) and retrieves coefficients for linear evaluation (0032; fig. 1B). Mangnall teaches bit-based table indexing (0041, 0051 and denser table entries where greater curvature increases interpolation error (0043-0045; Fig. 7). Applying Mangnall’s indexing and variable spacing to Wang’s translated input evaluator would therefore address both input-resolution error and interpolation error. Their shared use of floating-point inputs, lookup tables, and linear evaluation provides a technical basis for expecting a compatible implementation.
Additionally, it would have been obvious for an ordinary skill to consider Bar-on to construct the approximation segments and reduce coefficient-storage and locator complexity. Bar-on defines segments boundaries and fits coefficients using approximation-error measures (0034-0037) and expressly considers the smallest existing segment first during merging to reduce the segment locator’s bitmap width (0028; Figs. 2-5). Incorporating that construction procedure into the Wang-Mangnall evaluator would provide a method for balancing segment count, approximation accuracy, and indexing hardware. This rationale translating and input from a shortest final segment to zero while preserving floating-point bit-defined segmentation. By this rationale, claim 1 is rejected.
Regarding claims 4, 8-10, 14, 16-17, and 20, the combination Wang-Mangnall/Bar-On teaches:
4. The method of claim 1, wherein the approximating the plurality of segments comprises: identifying and converting a sample input in the floating point format in an input range of the nonlinear function; identifying a segment including the sample input; calculating an output of the nonlinear function corresponding to the sample input; deriving, based on a plurality of sample inputs included in one segment and a plurality of outputs respectively corresponding to the plurality of sample inputs, a linear function corresponding to the one segment; and adding at least one coefficient of the linear function to a lookup table (Bar-On 0035-0037, 0039-0042, 0095-0099; Fig. 7: sample-based error minimization and coefficient generation/storage. Apply Wang’s 0026-0028) conversion to the samples).
8. The method of claim 1, wherein the nonlinear function includes an activation function of an artificial neural network (Mangnall, 0010-0011, 0037; Figs. 1, 1A, and Bar-On 0032, 0044).
9. An apparatus comprising: a memory that stores a series of instructions; and at least one processor configured to perform the method of claim 1 by executing the series of instructions (Mangnall, 0023, 0037-0040; Fig. 2; Wang, 0093; Bar-On 0095-0099; Fig. 7).
10. A method of generating an output of a nonlinear function approximated in units of a plurality of segments, the method comprising: generating a converted input by subtracting a constant from an input in a floating point format (Wang, 0026-0027, 0090; Fig. 1A, and 8); identifying a segment, including the converted input, among the plurality of segments based on at least one bit of the converted input (Wang, 0091-0092; Fig. 8, combined with Mangall, 0041, 0051; and Figs. 3 and 6); obtaining at least one coefficient of a linear function corresponding to the segment, that is identified, from a lookup table (Wang, 0032, 0089, 0092; Figs 1B, and 8); and calculating the output of the linear function based on the converted input and the at least one coefficient (Bar-On 0035-0037, 0039-0042, 0095-0099; Fig. 7: sample-based error minimization and coefficient generation/storage. Apply Wang’s 0026-0028), wherein the constant corresponds to an amount by which the nonlinear function is shifted so that a shortest segment among the plurality of segments includes zero in the floating point format (Mangall, 0044-0045; see also Bar-On’s smallest segment selection concerns merging). The same motivation and reason to combine used for the rejection of claim 1 is also valid for this claim.
14. The method of claim 10, wherein the nonlinear function includes an activation function of an artificial neural network (Mangnall, 0010-0011, 0037; Figs. 1, 1A, and Bar-On 0032, 0044).
16. An apparatus comprising: a memory that stores a series of instructions; and at least one processor configured to perform the method of claim 10 by executing the series of instructions (Mangnall, 0023, 0037-0040; Fig. 2; Wang, 0093; Bar-On 0095-0099; Fig. 7).
17. An apparatus for generating an output of a nonlinear function approximated in units of a plurality of segments, the apparatus comprising: a floating point adder configured to generate a converted input by subtracting a constant from an input in a floating point format (Wang, 0026-0027, 0090; Fig. 1A, and 8); an index generator configured to generate an index of a segment, including the converted input, among the plurality of segments based on at least one bit of the converted input (Wang, 0091-0092; Fig. 8, combined with Mangall, 0041, 0051; and Figs. 3 and 6); and a linear calculator configured to obtain at least one coefficient of a linear function based on the index from a lookup table and generate the output based on the input and the at least one coefficient (Bar-On 0035-0037, 0039-0042, 0095-0099; Fig. 7: sample-based error minimization and coefficient generation/storage. Apply Wang’s 0026-0028), wherein the constant corresponds to an amount by which the nonlinear function is shifted so that a shortest segment among the plurality of segments includes zero in the floating point format (Mangall, 0044-0045; see also Bar-On’s smallest segment selection concerns merging). The same motivation and reason to combine used for the rejection of claim 1 is also valid for this claim.
20. The apparatus of claim 17, wherein the nonlinear function comprises an activation function of an artificial neural network (Mangnall, 0010-0011, 0037; Figs. 1, 1A, and Bar-On 0032, 0044).
Allowable Subject Matter
Claims 2-3, 5-7, 11-13, 15, and 19 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.
CONCLUSION
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jude Jean-Gilles whose telephone number is 571-272-3914. The examiner can normally be reached on Mon-Fri, from 9:00AM-5:00PM.
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, Tonia Dollinger can be reached on 571-272-4170. 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.
/JUDE JEAN GILLES/Primary Examiner, Art Unit 2459
October 26, 2026