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 .
Status
This instant application No. 19/112,411 has claims 1-7 pending based on the preliminary amendment filed on March 17, 2025.
Priority / Filing Date
Applicant’s claim for priority of PCT/JP2022/038258 (filed on October 13, 2022) is acknowledged. The effective filing date for this application is October 13, 2022.
Title
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed. The title fails to accurately reflect the scope of the claims since overly broad, vague, or generic terms (i.e., information processing) that do not convey the actual invention.
Drawings
The drawings filed on March 17, 2025 are acceptable for examination purposes.
Information Disclosure Statement
As required by M.P.E.P. 609(C), the Applicant’s submission of the Information Disclosure Statement filed on March 17, 2025 is acknowledged by the Examiner and the cited references have been considered in the examination of the claims now pending. As required by M.P.E.P. 609 C(2), a copy of the PTOL-1449s initialed and dated by the Examiner is attached to the instant Office action.
Claim Rejections - 35 USC § 112
35 U.S.C. 112 reads as follows:
(b) 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 1-7 are rejected under 35 U.S.C. 112(b) for failing to particularly point out and distinctly claim the subject matter.
Claims 1, and 6-7 recite in each claim the indefinite term milder. Since the term is not recognized term of art in machine learning or mathematics for describing activation functions. It is unclear if this strictly requires the absolute value of the gradient to be less than 1 across the entire domain of the function, or if it applies only to specific regions, averages, or asymptotic limits. Without an objective standard, explicit mathematical definition, or benchmark clarifying exactly how to measure when a change becomes milder, the term renders the metes and bounds of the claim subjective.
Claims 2-5 fail to resolve the deficiencies in claim 1 since they only further limit claim 1. Thus, claims 2-5 are also rejected for the same reasons presented above.
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.
The claimed invention in claims 1-7 are directed to a judicial exception (i.e., an abstract idea) without significantly more.
Claims 1-7 pass step 1 of the 35 U.S.C. 101 analysis since each claim is directed to a method; an apparatus comprising at least one memory and at least one processor (i.e., hardware components per Figure 1 as known in the art); or a non-transitory recording medium.
a. Claims 1, and 6-7 recite in each claim, in part, elements that are directed to an abstract idea (“Courts have examined claims that required the use of a computer and still found that the underlying, patent-ineligible invention could be performed via pen and paper or in a person’s mind.” Versata Dev. Group v. SAP Am., Inc., 793 F.3d 1306, 1335, 115 USPQ2d 1681, 1702 (Fed. Cir. 2015)). Each claim recites acquiring series data, calculating a likelihood ratio using an activation function defined by specific value range and gradient rules, and classifying the data based on this ratio. These limitations are directed to a mathematical concept of calculating a likelihood ratio via mathematical relationships, and a mental process of classifying data into categories per Step 2A – Prong 1 of the abstract idea analysis.
Per Step 2A – Prong 2, each claim does not include additional elements that integrate the abstract idea into a practical application. The recitation of at least one memory…to execute the instructions, method that is executed by at least one computer, and non-transitory recording medium merely invokes generic computer components as a tool to perform the mathematical and classification steps. Each claim fails to effect an improvement in computer functionality or apply the exception to a specific technological environment.
Per Step 2B, the additional elements, considered both individually and as an ordered combination, are well-understood, routine, and conventional (WURC) computer hardware components. They are recited at a high level of generality and do not provide an inventive concept sufficient to transform the abstract idea into a patent eligible application. Thus, the claims are ineligible under 35 U.S.C. 101.
b. Claims 2-3 further define the activation function as a combination or square root functions arranged to be point-symmetrical around the origin, and narrow this down to a specific mathematical formula. These limitations explicitly recite mathematical formulas and calculations, which are abstract ideas. These claims merely provide the specific mathematical algorithms used to perform the abstract idea in the respective independent claim. They do not add any additional hardware or practical application steps to integrate the mathematical concepts into a practical application, nor do they add an inventive concept beyond the WURC components recited in claim 1. Thus, the claims are ineligible under 35 U.S.C. 101.
c. Claims 4-5 further define the activation function as a combination of log functions arranged to be point-symmetrical around the origin., and specifying a mathematical formula to do as such. These limitations explicitly recite mathematical relationships and formulas which fail squarely within the judicial exception of a mathematical concept. Relying on WURC components of a generic computer fails to integrate the specific logarithmic formulas into a practical application or provide an inventive concept. The claims amount to nothing more than executing a mathematical algorithm on a generic computer. Thus, the claims are ineligible under 35 U.S.C. 101.
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 1-2, and 6-7 are rejected under AIA 35 U.S.C. 103 as being unpatentable over Stolpman et al. (Pub. No. US 2018/0137086, published on May 17, 2019; hereinafter Stolpman) in view of Parisi (“m-arcsinh: An Efficient and Reliable Function for SVM and MLP in scikit-learn”, published on September 16, 2020).
Regarding claims 1, and 6-7 Stolpman clearly shows and discloses an information processing method that is executed by at least one computer, the information processing method; an information processing apparatus comprising: at least one memory that is configured to store instructions; and at least one processor that is configured to execute the instructions to implement the method; and a non-transitory recording medium on which a computer program that allows at least one computer to execute an information processing method is recorded, the information processing method (Abstract and Figure 3) including:
acquiring a plurality of elements included in series data (The multiple measurement data samples can include a sequence of measurement data samples collected from at least one sensor for detecting physical items in a moveable platform, [0018]. The process of FIG. 3 includes receiving (at 302) measurement data from at least one sensor (e.g., the ToF sensor 116) that detects a signal reflected from a surface inside the moveable platform, [0046]. The received discrete I&Q samples, [0080]-[0081]);
calculating a likelihood ratio indicating likelihood of a class to which the series data belong (a likelihood ratio test generally refers to a statistical test that is used to compare the goodness of fit of two models, [0017]. The load status detector 208 includes a likelihood ratio test (LRT) function 209 that is used to apply a likelihood ratio test using measurement data from the ToF sensor 116 for determining the load status of the container 104, [0038]. The SPRT sums or otherwise aggregates a number of LLR output values to produce an SPRT output value, and the SPRT output value, [0100]-[0101]), from at least two consecutive elements out of the plurality of elements (with each including the latest (n+1) samples, [0102]-[0103]. It is clear that the sequence contains the latest n + 1 samples and the SPRT value is obtained by summing the LLR values for those samples. For N > 1, the calculation necessarily uses at least two successive samples); and
classifying the series data into at least one class of a plurality of classes serving as classification candidates, based on the likelihood ratio (the processor 202 receives measurement data from the ToF sensor 116, and based on the measurement data, the processor 202 is able to determine a load status of the container 104 (either the load status of the entirety of the container, or the load status of a zone of multiple zones in the container), [0034]).
Parisi then discloses calculating a likelihood ratio indicating likelihood of a class to which the series data belong, by using an activation function in which a value range is wider than [-1,1] (the disclosed function includes a factor that grows with
√
|
x
|
and arcsinh(x) factor. Because arcsinh(x) grows logarithmically for large |x|, the product grows without bound as |x| increases. Thus, the output is not confined to the interval [-1, 1], [Section 2.3]), in which a change in an output value is milder than a change in an input value (The m-arcsinh function also changes more mildly than its input for sufficiently large input magnitudes. Specifically, its asymptotic growth is proportional to
x
log
x
which is sublinear relative to |x|. Consequently, the function compresses the magnitude of large inputs rather than reproducing the input with a one-to-one linear relationship, [Section 2.3, 4 & 5]), and in which a gradient at an origin does not diverge (the mathematical definition of m-arcsinh and its derivative evaluated at the origin using the definition of a derivative yields f’(0)=0. As such, the claimed non-diverging gradient property is necessarily present by the activation function disclosed in [Section 2.3]).
It would have been obvious to an ordinary person skilled in the art at the time of the invention was effectively filed to incorporate the teachings of Parisi with the teachings of Stolpman for the purpose of providing nonlinear processing of measurement data features while avoiding a bounded output range and maintaining finite gradient behavior at the origin by applying a known activation function for classification related properties in a numerical classification system.
Regarding claim 2, Parisi further discloses the activation function is a function in which square root functions are combined so as to be point-symmetrical around the origin (the hyperbolic nature of the inverse hyperbolic sine function (’arcsinh’), suitable for MLP, and the slightly non-linear characteristic of the squared root function, appropriate for SVM, [Section 2.3]. It is clear that because arcsinh(x) is odd and
√
|
x
|
is even, the product is odd and therefore point-symmetric about the origin).
Claim 4 is rejected under AIA 35 U.S.C. 103 as being unpatentable over Stolpman in view of Parisi and further in view of Martin et al. (Pat. No. US 10698593, published on June 30, 2020; hereinafter Martin).
Regarding claim 4, Martin then discloses the activation function is a function in which log functions are combined so as to be point-symmetrical around the origin (FIG. 4 illustrates three types of axis scaling that can be used for data visualization axes. The first example is no scaling (402). The data values are not scaled at all. The second example is logarithmic scaling 404. In this case, a logarithmic transform is applied to values on the axis. As illustrated in FIG. 4, logarithmic scaling only works when the data values are positive. The third example is symmetric log scaling 406. In some implementations, an inverse hyperbolic sine a sin h(x)=ln(x+
x
^
2
+
1
) is used for symmetric log axis scaling, [Column 7, Lines 31-49]. It is clear that the resulting activation function is odd and therefore point-symmetric about the origin).
It would have been obvious to an ordinary person skilled in the art at the time of the invention was effectively filed to incorporate the teachings of Martin with the teachings of Stolpman, as modified by Parisi, for the purpose of providing nonlinear processing of measurement data features using inverse-hyperbolic-sine component in logarithmic form when performing the numerical classification.
Allowable Subject Matter
Claims 3 and 5 are each objected for being dependent on a rejected base claim but would be allowable over the prior art (given that all other pending rejections are resolved) if rewritten in independent form to incorporate all limitations of the base claim and all dependent claim(s).
Relevant Prior Art
The following references are deemed relevant to the claims:
Qiu et al. (Pub. No. US 2022/0122103) teaches on the basis of a depth auto-encoder, a neighborhood association method and a similarity difference compensation method, performing difference compensation correction on a calculation simulation data set by using a historical actual measurement data set; and training a BP neural network model by using the calculation simulation data set after the difference compensation correction to serve as a performance prediction method of a customized product. By combining the depth auto-encoder, and by utilizing the neighborhood association method and the similarity difference compensation method, low-fidelity calculation simulation data is associated with high-fidelity historical actual measurement data.
Zhang et al. (Pub. No. US 2020/0057175) teaches a real time data query and storage mechanism that integrates the fetched NWP data with historical PV generation data to train a PV forecasting engine. A hybrid PV forecasting process, using both persistent and long-short-term-memory neural network (LSTM-NN) training can be used as the training model of the PV forecasting engine. LSTM-NN architecture is more suitable for time series forecasting due to its recurrent architecture and memory units, while persistent model performs best for consecutive sunny day predictions. Using weather data from NWP model, the example embodiments provide a hybrid PV forecasting method that outperforms many other PV forecasting methods regarding prediction accuracy and robustness.
Contact Information
Any inquiry concerning this communication or earlier communications from the Examiner should be directed to Son Hoang whose telephone number is (571) 270-1752. The Examiner can normally be reached on Monday – Friday (7:00 AM – 4:00 PM).
If attempts to reach the Examiner by telephone are unsuccessful, the Examiner’s supervisor, Sherief Badawi can be reached on (571) 272-9782. 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.
/SON T HOANG/Primary Examiner, Art Unit 2169
September 4, 2026
Prosecution Insights