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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on February is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Objections
Claim 13 is objected to because of the following informalities: In the limitation “determining priorities of of the plurality of explanatory variables according to a condition detected to be satisfied”, “of” is repeated twice in line 7. Appropriate correction is required.
Claim Interpretation
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: processing unit, detection unit, determination unit, and learning unit in claims 1 and 12.
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 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 1-14 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.
Claim 1:
Claim 1 recites the limitation "defining timings to perform learning of a regression model configured to predict one or more objective variables for a plurality of explanatory variables are satisfied" in lines 4-6. The term “satisfied” in claim 1 is a relative term which renders the claim indefinite. The term “satisfied” 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. The limitation "determine priorities of the plurality of explanatory variables according to a condition detected to be satisfied." in lines 7-9 has also been rendered indefinite by the use of the term "satisfied".
Claims 2-12 inherit the deficiencies of claim 1 and are therefore also rejected as being indefinite.
Claim 12:
Claim 12 recites the limitation " " in lines 3-4. The term “satisfied” in claim 12 is a relative term which renders the claim indefinite. The term “satisfied” 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 13:
Claim 13 recites the limitation "defining timings to perform learning of a regression model configured to predict one or more objective variables for a plurality of explanatory variables are satisfied" in lines 4-6. The term “satisfied” in claim 13 is a relative term which renders the claim indefinite. The term “satisfied” 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. T.
Claim 14:
Claim 14 recites the limitation "defining timings to perform learning of a regression model configured to predict one or more objective variables for a plurality of explanatory variables are satisfied" in lines 5-7. The term “satisfied” in claim 14 is a relative term which renders the claim indefinite. The term “satisfied” 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. T.
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 14 is rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter because there is no definition of computer readable medium in the applicant’s specification. Therefore, under BRI non-transient computer readable medium could include signals making the claim signals per se. As such the claim is rejected under failing to fall into the one of the statutory categories of the patent eligible subject matter.
Claims 1-14 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea (mathematical concept and mental process) without significantly more.
Claim 1:
Regarding claim 1, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites “An information processing apparatus”, and an apparatus or machine is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
“detect whether or not one or more conditions defining timings to perform learning of… predict one or more objective variables for a plurality of explanatory variables are satisfied;” (this is a mental process, a person could mentally evaluate whether or not one or more conditions defining timings to perform learning of a model to predict one or more objective variables for a plurality of explanatory variables are satisfied, see MPEP § 2106.04(a)(2)(III)),
“determine priorities of the plurality of explanatory variables according to a condition detected to be satisfied;” (this is a mental process, a person could mentally determine a priority of an explanatory variable according to a satisfied condition, see MPEP § 2106.04(a)(2)(III)),
“and perform learning of the regression model by using an objective function and learning data, the objective function including a regularization term having a regularization strength changing according to the priorities;” (this is a mathematical concept, performing learning of a model using an objective function with a regularization term having a regularization strength is a mathematical formula, see paragraphs [0068-0074] and equation 4 in Specification, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under the broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process and mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
“An information processing apparatus comprising a processing unit configured to:” (A processing apparatus comprising a processing unit is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)),
“…a regression model configured to…” (Configuring a regression model is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional element iv recites generic computer component being used as tool to perform functions of the judicial exception, and additional element v recites mere instructions to apply an exception using generic computer, which is not indicative of significantly more.
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claim 2:
Regarding claim 2, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 2 recites the following additional elements:
“The apparatus according to claim 1, wherein…predict a plurality of objective variables, and…calculate the plurality of explanatory variables by multiplying one or more first variables and a plurality of dummy variables corresponding to the plurality of objective variables,” (this is a mathematical concept, calculating a plurality of variables by multiplying one or more first variables and a plurality of dummy variables is a mathematical calculation, see MPEP § 2106.04(a)(2)(I)),
“and generate the learning data including the plurality of calculated explanatory variables;” (this is a mental process, a person can mentally evaluate learning data that includes calculated variables, see MPEP § 2106.04(a)(2)(III)),
“and perform learning of the regression model using the generated learning data.” (this is a mathematical concept, performing learning of a model is a mathematical process, see paragraphs [0068-0074], see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept and mental process but for the recitation of generic computer components, then it falls within the mathematical concept and mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“…the regression model is configured to...” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
“…the processing unit is configured to:…” (In step 2A, prong 2, this is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)). (In step 2B, this is also considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 3:
Regarding claim 3, it is dependent upon claim 2, and thereby incorporates the limitations of, and corresponding analysis to claim 2. Further, claim 3 recites the following additional elements:
“The apparatus according to claim 2, wherein the…specify a type of objective variable corresponding to a dummy variable as a type of explanatory variable, and generate first correspondence information in which the specified type of explanatory variable and the explanatory variable are associated with each other;” (this is a mental process, a person could mentally evaluate specifying a type of objective variable that corresponds with a dummy variable as an explanatory variable and evaluate generating information in which the specified type of explanatory variable and explanatory variable are associated, see MPEP § 2106.04(a)(2)(III)),
“and determine a priority corresponding to a type of the detected condition for the explanatory variable included in the type of explanatory variable corresponding to the type of the detected condition by using the first correspondence information and second correspondence information in which the type of condition, the type of explanatory variable, and the priority are associated with each other.” (this is a mental process, a person can mentally determine a priority corresponding to a type of a detected condition for an explanatory variable included in the type of explanatory variable corresponding to the type of the detected condition by using information and information in which the type of condition, the type of explanatory variable, and the priority are associated with each other., see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“…the processing unit is configured to:…” (In step 2A, prong 2, this is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)). (In step 2B, this is also considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 4:
Regarding claim 4, it is dependent upon claim 2, and thereby incorporates the limitations of, and corresponding analysis to claim 2. Further, claim 4 recites the following additional elements:
“The apparatus according to claim 2, wherein…determine a priority of an explanatory variable corresponding to an objective variable according to a magnitude of a prediction error by the regression model after learning among the plurality of objective variables.” (this is a mental process, a person could mentally determine a priority of an explanatory variable that corresponds to an objective variable, see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“…the processing unit is configured to:…” (In step 2A, prong 2, this is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)). (In step 2B, this is also considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 5:
Regarding claim 5, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 5 recites the following additional elements:
“The apparatus according to claim 1, wherein…determine a priority corresponding to the detected condition for an explanatory variable corresponding to the detected condition by using correspondence information in which the condition, the explanatory variable, and the priority are associated with each other.” (this is a mental process, a person could mentally determine a priority that corresponds to a detected condition for an explanatory variable that corresponds to the detected condition using information where the condition, variable and priority are associated, see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“…the processing unit is configured to…” (In step 2A, prong 2, this is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)). (In step 2B, this is also considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 6:
Regarding claim 6, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 6 recites the following additional elements:
“The apparatus according to claim 1, wherein…generate the learning data including one or more explanatory variables having priorities higher than another explanatory variable among the plurality of explanatory variables, and one or more objective variables;” (this is a mental process, a person can mentally evaluate generating data that includes explanatory variables having a priority higher than another explanatory variable from the plurality of explanatory variables, and objective variables, see MPEP § 2106.04(a)(2)(III)),
“and perform learning of the regression model using the generated learning data.” (this is a mathematical concept, performing learning of a model is a mathematical process, see paragraphs [0068-0074], see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“…the processing unit is configured to:…” (In step 2A, prong 2, this is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)). (In step 2B, this is also considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 7:
Regarding claim 7, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 7 recites the following additional elements:
“and determine the priorities of the plurality of explanatory variables according to magnitudes of the changes.” (this is a mental process, a person can mentally determine priorities of a plurality of explanatory variables according to a magnitude of changes, see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“The apparatus according to claim 1, wherein the processing unit is configured to obtain changes of the plurality of explanatory variables between the learning data,” (In step 2A, prong 2, this is considered insignificant extra-solution activity of mere data gathering – see MPEP § 2106.05(g)). (In step 2B, this is also considered insignificant extra-solution activity of mere data gathering, which is a well understood routine and conventional activity, see receiving or transmitting data over a network, e.g., using the Internet to gather data, Symantec, 838 F.3d at 1321, 120 USPQ2d at 1362 – see MPEP § 2106.05(g)).
“and test data serving as an input in prediction using the regression model after learning,” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 8:
Regarding claim 8, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 8 recites the following additional elements:
“The apparatus according to claim 1, wherein the priorities include selection priorities representing priorities of selecting the plurality of explanatory variables and update priorities representing priorities of updating the plurality of explanatory variables,” (this is a mental process, a person could mentally evaluate the priorities including selection priorities that represent priorities of selecting a plurality of explanatory variables and evaluate updating priorities that represent priorities of updating a plurality of explanatory variables, see MPEP § 2106.04(a)(2)(III)),
“and the objective function includes: a term evaluating compatibility between a prediction result and correct data;” (this is a mathematical concept, an objective function including a term for evaluating compatibility between prediction results and correct data is a mathematical formula, see paragraph [0070] and equation 4 in Specification, see MPEP § 2106.04(a)(2)(I)),
“a first regularization term obtained by multiplying terms for regularizing parameters corresponding to the plurality of explanatory variables by weights based on the selection priorities;” (this is a mathematical concept, a regularization term obtained by multiplying terms to regularize parameters is a mathematical calculation, see MPEP § 2106.04(a)(2)(I)),
“and a second regularization term obtained by multiplying terms for regularizing changes with respect to the parameters before update by weights based on the update priorities.” (this is a mathematical concept, a regularization term obtained by multiplying terms to regularize changes with respect to parameters is a mathematical calculation, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept and mental process but for the recitation of generic computer components, then it falls within the mathematical concept and mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 9:
Regarding claim 9, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 9 recites the following additional elements:
“The apparatus according to claim 1, wherein…predict the one or more objective variables for test data including the plurality of explanatory variables...” (this is a mental process, a person could mentally predict variables for test data that includes a plurality of explanatory variables, see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“…the processing unit is configured to:…” (In step 2A, prong 2, this is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)). (In step 2B, this is also considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)).
“…using the regression model after learning.” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 10:
Regarding claim 10, it is dependent upon claim 9, and thereby incorporates the limitations of, and corresponding analysis to claim 9. Further, claim 10 recites the following additional elements:
“The apparatus according to claim 9, wherein…estimate whether or not an object related to the test data is in a specific state based on a prediction error of the predicted objective variables.” (this is a mental process, a person can mentally estimate whether or not on object that is related to test data is in a specific state based on a prediction error of predicted objective variables, see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“…the processing unit is configured to:…” (In step 2A, prong 2, this is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)). (In step 2B, this is also considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 11:
Regarding claim 11, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 11 recites the following additional elements:
“perform learning of the regression model using the learning data so as to optimize the objective function..” (this is a mathematical concept, performing learning of a model is a mathematical process, see paragraphs [0068-0074], see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“The apparatus according to claim 1, wherein the processing unit is configured to…” (In step 2A, prong 2, this is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)). (In step 2B, this is also considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 12:
Regarding claim 12, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 12 recites the following additional elements:
“…detect whether or not the one or more conditions are satisfied;” (this is a mental process, a person could mentally evaluate detecting if conditions are satisfied, see MPEP § 2106.04(a)(2)(III)),
“…determine the priorities;” (this is a mental process, a person can mentally determine priorities, see MPEP § 2106.04(a)(2)(III)),
“…perform learning of the regression model.” (this is a mathematical concept, performing learning of a model is a mathematical process, see paragraphs [0068-0074], see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
“The apparatus according to claim 1, wherein the processing unit comprises: a detection unit configured to...” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
“and a determination unit configured to...” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
“and a learning unit configured to...” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 13:
Regarding claim 13, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites “An information processing method executed by an information processing apparatus, the method comprising:”, and a method or process is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
“detecting whether or not one or more conditions defining timings to perform learning of…predicts one or more objective variables for a plurality of explanatory variables are satisfied;” (this is a mental process, a person could mentally evaluate whether or not one or more conditions defining timings to perform learning of a model to predict one or more objective variables for a plurality of explanatory variables are satisfied, see MPEP § 2106.04(a)(2)(III)),
“determining priorities of of the plurality of explanatory variables according to a condition detected to be satisfied;” (this is a mental process, a person could mentally determine a priority of an explanatory variable according to a satisfied condition, see MPEP § 2106.04(a)(2)(III)),
“and performing learning of the regression model by using an objective function and learning data, the objective function including a regularization term having a regularization strength changing according to the priorities.” (this is a mathematical concept, performing learning of a model using an objective function with a regularization term having a regularization strength is a mathematical formula, see paragraphs [0068-0074] and equation 4 in Specification, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under the broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process and mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
“An information processing method executed by an information processing apparatus, the method comprising:” (A processing apparatus is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)),
“…a regression model that…” (Configuring a regression model is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional element iv recites generic computer component being used as tool to perform functions of the judicial exception, and additional element v recites mere instructions to apply an exception using generic computer, which is not indicative of significantly more.
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claim 14:
Regarding claim 14, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites “A computer program product comprising a computer-readable medium including programmed instructions, the instructions causing a computer to execute:”, and a computer-readable medium or manufacture is not of the four statutory categories of invention because there is no definition of computer readable medium in the applicant’s specification. Therefore, under BRI computer readable medium could include signals making the claim signals per se.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
“detecting whether or not one or more conditions defining timings to perform learning of…predicts one or more objective variables for a plurality of explanatory variables are satisfied;” (this is a mental process, a person could mentally evaluate whether or not one or more conditions defining timings to perform learning of a model to predict one or more objective variables for a plurality of explanatory variables are satisfied, see MPEP § 2106.04(a)(2)(III)),
“determining priorities of the plurality of explanatory variables according to a condition detected to be satisfied;” (this is a mental process, a person could mentally determine a priority of an explanatory variable according to a satisfied condition, see MPEP § 2106.04(a)(2)(III)),
“and performing learning of the regression model by using an objective function and learning data, the objective function including a regularization term having a regularization strength changing according to the priorities.” (this is a mathematical concept, an objective function with a regularization term having a regularization strength is a mathematical formula, see paragraph [0070] and equation 4 in Specification, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under the broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process and mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
“A computer program product comprising a computer-readable medium including programmed instructions, the instructions causing a computer to execute:” (A processing computer-readable medium is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)),
“…a regression model that…” (Configuring a regression model is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional element iv recites generic computer component being used as tool to perform functions of the judicial exception, and additional element v recites mere instructions to apply an exception using generic computer, which is not indicative of significantly more.
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
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.
Claim(s) 1, 5, 9-11, and 13-14 are rejected under 35 U.S.C. 103 as being unpatentable over Chakravorty J. et al, (US. Patent Application Publication 20240112073 A1) effectively filed on September 22, 2022, (hereafter Chakravorty), in view of Watanabe T. et al, (US. Patent Application Publication 20190369572 A1) effectively filed on March 11, 2019, and further in view of Szmulewicz D. et al, (US. Patent Application Publication 20230337944 A1) effectively filed on June 18, 2021.
Claim 1:
Regarding claim 1, Chakravorty teaches “An information processing apparatus comprising a processing unit configured to: detect whether or not one or more conditions defining timings to perform learning of a regression model configured to predict one or more objective variables for a plurality of explanatory variables are satisfied;”
See Chakravorty in paragraph [0016] describing, “In addition, any of the methods, described above and elsewhere herein, may be embodied, individually or in any combination, in executable software modules of a processor-based system, such as a server, and/or in executable instructions stored in a non-transitory computer-readable medium.” Here, Chakravorty establishes the embodiment described being a method with a system or apparatus with a processor which is being interpreted as the processing unit, and a computer-readable medium which is being interpreted as the computer program product. Further, see Chakravorty in paragraph [0100] describing, “In subprocess 433, the parameters of machine-learning model 334 are determined and the error of the optimization problem is calculated. The parameters of machine-learning model 334 may comprise the weights and/or other variables in machine-learning model 334 (i.e., ƒθ). It should be understood that these weights are different than the weights λ assigned to the clusters in subprocess 432.” Here, Chakravorty establishes a subprocess that shows a model with various variables which are being interpreted as the explanatory variables. Further, see Chakravorty in paragraph [0101] describing, “In subprocess 434, it is determined whether or not a first stopping condition for the inner loop is satisfied. The first stopping condition may comprise a number of iterations (i.e., training epochs) of the inner loop reaching or exceeding a threshold, a change in the error of the optimization problem from the preceding iteration reaching or falling below a threshold, an error of the optimization problem reaching or falling below a threshold, a time duration of the inner loop reaching or exceeding a threshold, and/or the like. If the first stopping condition has been satisfied (i.e., “Yes” in subprocess 434), training subprocess 430 proceeds to subprocess 435.” Here, Chakravorty establishes detecting of a condition here for the learning of a model with the training. Further, see Chakravorty in paragraph [0118] describing, “Experiments were conducted using machine-learning model 334 with Laplace regularization for the task of fault localization in time domain protection. In particular, a regression model for estimating fault location was trained with Laplace regularization, using process 400, and without Laplace regularization.” Here, Chakravorty establishes the model for these processes as a regression model for estimating fault location which is the same as being configured to predict with the estimating, and the fault location can be the objective variable.
Further, Chakravorty teaches “and perform learning of the regression model by using an objective function and learning data,...”
See Chakravorty in paragraph [0124] describing, “Disclosed embodiments frame the training of a machine-learning model 334 as a multi-objective optimization problem. In addition to the primary objective function that calculates an error of machine-learning model 334 (e.g., as a regression task), the multi-objective optimization problem comprises a non-smoothness penalization function gθ, which drives the primary objective function to a solution that respects smoothness between the inputs and outputs of machine-learning model 334.” Here, Chakravorty establishes performing learning of the regression model with the training of the machine learning model established to be a regression model in previous limitation using a multi-objective optimization problem containing a primary objective function that comprises of a non-smoothness penalization function. Further, see Chakravorty in paragraph [0125] describing, “This Laplace regularization, using non-smoothness penalization function gθ, with or without adaptive weights for clusters, can be employed in any regression task in which the data are heterogeneous and the regression function is foreseen to be smooth on the data (i.e., the regression function does not vary abruptly with the data points).” Here, Chakravorty establishes Laplace regularization, which is known in the art to be a regularization term in itself, with the non-smoothness penalization function using data and the function with adaptive weights which can be seen as the regularization strength being changed according to priorities as the weight change according to priorities was established in previous limitations.
Chakravorty does not appear to explicitly teach “determine priorities of the plurality of explanatory variables according to a condition detected to be satisfied; …the objective function including a regularization term having a regularization strength changing according to the priorities.”,
However in the same field of art, Watanabe teaches “determine priorities of the plurality of explanatory variables according to a condition detected to be satisfied;”
See Watanabe in paragraph [0087] describing, “Specifically, the regression model generator 71 uses the generally known Random forest algorithm to calculate indexes of priority P for each of weather elements, demand factors, and data demand densities available as explanatory variables and normalizes the indexes to 0 through 1. Suppose the normalized value of priority P for each of the weather elements, the demand factors, and the data demand densities satisfies “the condition of the explanatory variable used for the regression model” notified from the reference point adjustment processor 70.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Chakravorty with the teachings of Watanabe by using Chakravorty’s teachings of regularization techniques in regression model, and incorporate with Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models.
One of ordinary skill in the art would be motivated to do so because by integrating Watanabe’s frameworks into the methods of Chakravorty, which are both in the same field of art of regression models, one of ordinary skill in the art would bring a “an energy operation apparatus [that] operates energy in a management area based on a prediction result for supply and/or demand for the energy in the management area. The energy operation apparatus includes a demand predictor, a planner, an evaluator, and a solution quality controller. The demand predictor predicts demand and/or a power generation amount of future energy in a management area. The planner prepares a future energy supply plan in the management area based on a prediction result from the demand predictor. The evaluator evaluates supply and/or demand conditions including at least one of future weather in the management area, energy demand in the management area, a demand density and/or a generation density of future energy in the management area. The solution quality controller controls the quality of at least one of a prediction solution of the demand predictor and the energy supply plan of the planner based on an evaluation result from the evaluator.” (Watanabe, paragraph [0009]).
Neither Chakravorty or Watanabe appear to explicitly teach “…the objective function including a regularization term having a regularization strength changing according to the priorities.”,
However in the same field of art, Szmulewicz teaches “…the objective function including a regularization term having a regularization strength changing according to the priorities.”
See Szmulewicz in paragraph [0104] describing, “Random Subset Feature Selection (RSFS), Statistical Dependency (SDe) and Neighbourhood Component Analysis with regularization (NCA-R) were compared. NCA-R was selected because its low cross-validation error resulted in significantly fewer features. NCA-R learn the weights of features by minimising an objective function which measures the average leave-one-out (LOO) classification or regression loss over the training data and fine-tuning a regularisation parameter ‘λ’ to 0.0329 while evaluating the weights of the features such that the significance is indicative of the rank.” Here, Szmulewicz teaches an objective function with a regularization parameter or term that evaluates weights or strength that change significance based on the rank or priority.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty and Watanabe with the teachings of Szmulewicz by using Chakravorty’s teachings of regularization techniques in regression model and Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, and incorporate with Szmulewicz’s teachings of an objective function with a regularization term.
One of ordinary skill in the art would be motivated to do so because by integrating Szmulewicz’s frameworks into the methods of Chakravorty and Watanabe, one of ordinary skill in the art would bring a “a second algorithmic model used to calculate a severity score which is indicative of severity of the motor control disorder in the subject, and the machine learning approach is a random-forest regression model (RFR).” (Szmulewicz, paragraph [0040]).
Claim 5:
Regarding claim 5, Chakravorty in view of Watanabe, and further in view of Szmulewicz teaches the limitations of claim 1.
Chakravorty does not appear to explicitly teach “The apparatus according to claim 1, wherein the processing unit is configured to determine a priority corresponding to the detected condition for an explanatory variable corresponding to the detected condition by using correspondence information in which the condition, the explanatory variable, and the priority are associated with each other.”,
Further, Watanabe teaches “The apparatus according to claim 1, wherein the processing unit is configured to determine a priority corresponding to the detected condition for an explanatory variable corresponding to the detected condition by using correspondence information in which the condition, the explanatory variable, and the priority are associated with each other.”
See Watanabe in paragraph [0084] describing, “Actually, the reference point adjustment processor 70 maintains a second table 73 as illustrated in FIG. 8. The second table 73 defines types of the regression model to be generated by the regression model generator 71 and conditions of an explanatory variable (each weather element, each demand factor, and/or data demand density) used for the regression model corresponding to values of the predictive solution target stabilization index specified by the solution quality controller 36. A user previously creates the second table 73. In this case, the condition of the explanatory variable used for the regression model is provided as a threshold value (the “threshold” column in FIG. 8) of priority P for the explanatory variable.” Further, see Watanabe in paragraph [0085] describing, “The condition of the explanatory variable corresponds to the priority value of the explanatory variable. The reference point adjustment processor 70 selects and outputs the regression model and the condition of the explanatory variable used for the regression model to the regression model generator 71.” Further, see Watanabe in paragraph [0087] describing, “Specifically, the regression model generator 71 uses the generally known Random forest algorithm to calculate indexes of priority P for each of weather elements, demand factors, and data demand densities available as explanatory variables and normalizes the indexes to 0 through 1. Suppose the normalized value of priority P for each of the weather elements, the demand factors, and the data demand densities satisfies “the condition of the explanatory variable used for the regression model” notified from the reference point adjustment processor 70. Then, the regression model generator 71 uses these weather elements, demand factors, and data demand densities as explanatory variables to generate the regression model notified from the reference point adjustment processor 70.”Here, Watanabe establishes a priority of a condition that is determined with a threshold value to that corresponds to the explanatory variable. The type of condition and explanatory variable corresponds to each other and examples of the types are given with weather elements for example. The second table established with this information can be seen as the correspondence information, the condition, variable, and priority are all explicitly associated with each other.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Chakravorty with the teachings of Watanabe by using Chakravorty’s teachings of regularization techniques in regression model, and incorporate with Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models.
One of ordinary skill in the art would be motivated to do so because by integrating Watanabe’s frameworks into the methods of Chakravorty, which are both in the same field of art of regression models, one of ordinary skill in the art would bring a “an energy operation apparatus [that] operates energy in a management area based on a prediction result for supply and/or demand for the energy in the management area. The energy operation apparatus includes a demand predictor, a planner, an evaluator, and a solution quality controller. The demand predictor predicts demand and/or a power generation amount of future energy in a management area. The planner prepares a future energy supply plan in the management area based on a prediction result from the demand predictor. The evaluator evaluates supply and/or demand conditions including at least one of future weather in the management area, energy demand in the management area, a demand density and/or a generation density of future energy in the management area. The solution quality controller controls the quality of at least one of a prediction solution of the demand predictor and the energy supply plan of the planner based on an evaluation result from the evaluator.” (Watanabe, paragraph [0009]).
Claim 9:
Regarding claim 9, Chakravorty in view of Watanabe, and further in view of Szmulewicz teaches the limitations of claim 1.
Chakravorty, does not appear to explicitly teach “The apparatus according to claim 1, wherein the processing unit is configured to predict the one or more objective variables for test data including the plurality of explanatory variables using the regression model after learning.”,
However in the same field of art, Watanabe teaches “The apparatus according to claim 1, wherein the processing unit is configured to predict the one or more objective variables for test data including the plurality of explanatory variables using the regression model after learning.”
See Watanabe in paragraph [0164] describing, “The regression model generator 71 generates a regression model that satisfies the type and the condition of the regression model supplied from the reference point adjustment processor 70 and is found between the energy demand in the management area for the relevant energy operation apparatus 10 and an explanatory variable (S61). The regression model generator 71 outputs the generated regression model to the regressive prediction processor 72 (FIG. 6).” Further, see Watanabe in paragraph [0165] describing, “The regressive prediction processor 72 acquires a value of the necessary explanatory variable from the reference data storer 30 and a simulation result supplied from the physical simulator 63 of the second predictor 32. The regressive prediction processor 72 applies the acquired value to the regression model supplied from the regression model generator 71 and thereby calculates a predicted value for the energy demand at the time slice (specified time) in the management area of the relevant energy operation apparatus 10. The regressive prediction processor 72 outputs the calculated predicted value for the energy demand to the planning processor 101 (FIG. 6) of the planner 37 (FIG. 6) (S62).” Further, see Watanabe in paragraph [0133] describing, “The third predictor 33 predicts the future energy demand in the management area based on demand data stored in the demand data storer 40 (FIG. 6) of the reference data storer 30 (FIG. 6),”Here, Watanabe establishes a prediction of energy demand being interpreted as the objective variable here for management area which is from demand data stored in the storer which is being interpreted as the test data. Explanatory variables are included in the data storer that stores the demand data, and a regression model is used to predict the variables after learning by acquiring the data.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty with the teachings of Watanabe by using Chakravorty’s teachings of regularization techniques in regression model, and incorporate with Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models.
One of ordinary skill in the art would be motivated to do so because by integrating Watanabe’s frameworks into the methods of Chakravorty, which are both in the same field of art of regression models, one of ordinary skill in the art would bring a “an energy operation apparatus [that] operates energy in a management area based on a prediction result for supply and/or demand for the energy in the management area. The energy operation apparatus includes a demand predictor, a planner, an evaluator, and a solution quality controller. The demand predictor predicts demand and/or a power generation amount of future energy in a management area. The planner prepares a future energy supply plan in the management area based on a prediction result from the demand predictor. The evaluator evaluates supply and/or demand conditions including at least one of future weather in the management area, energy demand in the management area, a demand density and/or a generation density of future energy in the management area. The solution quality controller controls the quality of at least one of a prediction solution of the demand predictor and the energy supply plan of the planner based on an evaluation result from the evaluator.” (Watanabe, paragraph [0009]).
Claim 10:
Regarding claim 10, Chakravorty in view of Watanabe, and further in view of Szmulewicz teaches the limitations of claim 9.
Chakravorty does not appear to explicitly teach “The apparatus according to claim 9, wherein the processing unit is configured to estimate whether or not an object related to the test data is in a specific state based on a prediction error of the predicted objective variables.”,
However in the same field of art, Watanabe teaches “The apparatus according to claim 9, wherein the processing unit is configured to estimate whether or not an object related to the test data is in a specific state based on a prediction error of the predicted objective variables.”
See Watanabe in paragraph [0095] describing, “The temporal data distribution profile evaluator 80 is a functional part that is implemented when the CPU 11 executes a temporal data distribution profile evaluation program 80P (FIG. 5) stored in the external storage unit 13. The physical simulator 55 of the first predictor 31 performs the physical simulation to provide simulation results that are used as prediction results of weather data at reference points. The temporal data distribution profile evaluator 80 uses, for example, a dispersion state (dispersion value) of the predicted values to evaluate the magnitude of the temporal change in the prediction results and calculate a weighted average.” Here, Watanabe establishes predicted objective variables with the prediction results of weather data. An estimate of whether or not an object related to test data is in a specific state based on a prediction error is established here with the dispersion state of predicted values as the specific state based on prediction, and the error comes from the magnitude of temporal change, the weather data is established to be considered as the test data as it is used to perform physical simulation.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty with the teachings of Watanabe by using Chakravorty’s teachings of regularization techniques in regression model, and incorporate with Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models.
One of ordinary skill in the art would be motivated to do so because by integrating Watanabe’s frameworks into the methods of Chakravorty, which are both in the same field of art of regression models, one of ordinary skill in the art would bring a “an energy operation apparatus [that] operates energy in a management area based on a prediction result for supply and/or demand for the energy in the management area. The energy operation apparatus includes a demand predictor, a planner, an evaluator, and a solution quality controller. The demand predictor predicts demand and/or a power generation amount of future energy in a management area. The planner prepares a future energy supply plan in the management area based on a prediction result from the demand predictor. The evaluator evaluates supply and/or demand conditions including at least one of future weather in the management area, energy demand in the management area, a demand density and/or a generation density of future energy in the management area. The solution quality controller controls the quality of at least one of a prediction solution of the demand predictor and the energy supply plan of the planner based on an evaluation result from the evaluator.” (Watanabe, paragraph [0009]).
Claim 11:
Regarding claim 11, Chakravorty in view of Watanabe, and further in view of Szmulewicz teaches the limitations of claim 1.
Further, Chakravorty teaches ”The apparatus according to claim 1, wherein the processing unit is configured to perform learning of the regression model using the learning data so as to optimize the objective function.”
See Chakravorty in paragraph [0007] describing, “In an embodiment, a method of manifold regularization while training a machine-learning model, comprises using at least one hardware processor to: acquire a training dataset that comprises a plurality of feature sets, wherein each of the plurality of feature sets comprises a feature value for each of a plurality of features and is labeled with a target value for each of one or more targets; generate an optimization problem comprising an objective function and a non-smoothness penalization function, wherein the objective function calculates an estimated error between the target values and corresponding output values that the machine-learning model outputs for the plurality of feature sets, and wherein the non-smoothness penalization function is configured to increase the estimated error in the optimization problem as a smoothness of the machine-learning model decreases; and train the machine-learning model by adjusting the machine-learning model to minimize the estimated error, produced by the training dataset, in the optimization problem.”
Claim 13:
Regarding claim 13, Chakravorty teaches “An information processing method executed by an information processing apparatus, the method comprising: detecting whether or not one or more conditions defining timings to perform learning of a regression model that predicts one or more objective variables for a plurality of explanatory variables are satisfied;”
See Chakravorty in paragraph [0016] describing, “In addition, any of the methods, described above and elsewhere herein, may be embodied, individually or in any combination, in executable software modules of a processor-based system, such as a server, and/or in executable instructions stored in a non-transitory computer-readable medium.” Here, Chakravorty establishes the embodiment described being a method with a system or apparatus with a processor which is being interpreted as the processing unit, and a computer-readable medium which is being interpreted as the computer program product. Further, see Chakravorty in paragraph [0100] describing, “In subprocess 433, the parameters of machine-learning model 334 are determined and the error of the optimization problem is calculated. The parameters of machine-learning model 334 may comprise the weights and/or other variables in machine-learning model 334 (i.e., ƒθ). It should be understood that these weights are different than the weights λ assigned to the clusters in subprocess 432.” Here, Chakravorty establishes a subprocess that shows a model with various variables which are being interpreted as the explanatory variables. Further, see Chakravorty in paragraph [0101] describing, “In subprocess 434, it is determined whether or not a first stopping condition for the inner loop is satisfied. The first stopping condition may comprise a number of iterations (i.e., training epochs) of the inner loop reaching or exceeding a threshold, a change in the error of the optimization problem from the preceding iteration reaching or falling below a threshold, an error of the optimization problem reaching or falling below a threshold, a time duration of the inner loop reaching or exceeding a threshold, and/or the like. If the first stopping condition has been satisfied (i.e., “Yes” in subprocess 434), training subprocess 430 proceeds to subprocess 435.” Here, Chakravorty establishes detecting of a condition here for the learning of a model with the training. Further, see Chakravorty in paragraph [0118] describing, “Experiments were conducted using machine-learning model 334 with Laplace regularization for the task of fault localization in time domain protection. In particular, a regression model for estimating fault location was trained with Laplace regularization, using process 400, and without Laplace regularization.” Here, Chakravorty establishes the model for these processes as a regression model for estimating fault location which is the same as being configured to predict with the estimating, and the fault location can be the objective variable.
Further, Chakravorty teaches “and performing learning of the regression model by using an objective function and learning data,...”
See Chakravorty in paragraph [0124] describing, “Disclosed embodiments frame the training of a machine-learning model 334 as a multi-objective optimization problem. In addition to the primary objective function that calculates an error of machine-learning model 334 (e.g., as a regression task), the multi-objective optimization problem comprises a non-smoothness penalization function gθ, which drives the primary objective function to a solution that respects smoothness between the inputs and outputs of machine-learning model 334.” Here, Chakravorty establishes performing learning of the regression model with the training of the machine learning model established to be a regression model in previous limitation using a multi-objective optimization problem containing a primary objective function that comprises of a non-smoothness penalization function. Further, see Chakravorty in paragraph [0125] describing, “This Laplace regularization, using non-smoothness penalization function gθ, with or without adaptive weights for clusters, can be employed in any regression task in which the data are heterogeneous and the regression function is foreseen to be smooth on the data (i.e., the regression function does not vary abruptly with the data points).” Here, Chakravorty establishes Laplace regularization, which is known in the art to be a regularization term in itself, with the non-smoothness penalization function using data and the function with adaptive weights which can be seen as the regularization strength being changed according to priorities as the weight change according to priorities was established in previous limitations.
Chakravorty does not appear to explicitly teach “determining priorities of of the plurality of explanatory variables according to a condition detected to be satisfied; …the objective function including a regularization term having a regularization strength changing according to the priorities.”,
However in the same field of art, Watanabe teaches “determining priorities of of the plurality of explanatory variables according to a condition detected to be satisfied;”
See Watanabe in paragraph [0087] describing, “Specifically, the regression model generator 71 uses the generally known Random forest algorithm to calculate indexes of priority P for each of weather elements, demand factors, and data demand densities available as explanatory variables and normalizes the indexes to 0 through 1. Suppose the normalized value of priority P for each of the weather elements, the demand factors, and the data demand densities satisfies “the condition of the explanatory variable used for the regression model” notified from the reference point adjustment processor 70.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Chakravorty with the teachings of Watanabe by using Chakravorty’s teachings of regularization techniques in regression model, and incorporate with Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models.
One of ordinary skill in the art would be motivated to do so because by integrating Watanabe’s frameworks into the methods of Chakravorty, which are both in the same field of art of regression models, one of ordinary skill in the art would bring a “an energy operation apparatus [that] operates energy in a management area based on a prediction result for supply and/or demand for the energy in the management area. The energy operation apparatus includes a demand predictor, a planner, an evaluator, and a solution quality controller. The demand predictor predicts demand and/or a power generation amount of future energy in a management area. The planner prepares a future energy supply plan in the management area based on a prediction result from the demand predictor. The evaluator evaluates supply and/or demand conditions including at least one of future weather in the management area, energy demand in the management area, a demand density and/or a generation density of future energy in the management area. The solution quality controller controls the quality of at least one of a prediction solution of the demand predictor and the energy supply plan of the planner based on an evaluation result from the evaluator.” (Watanabe, paragraph [0009]).
Neither Chakravorty or Watanabe appear to explicitly teach “…the objective function including a regularization term having a regularization strength changing according to the priorities.”,
However in the same field of art, Szmulewicz teaches “…the objective function including a regularization term having a regularization strength changing according to the priorities.”
See Szmulewicz in paragraph [0104] describing, “Random Subset Feature Selection (RSFS), Statistical Dependency (SDe) and Neighbourhood Component Analysis with regularization (NCA-R) were compared. NCA-R was selected because its low cross-validation error resulted in significantly fewer features. NCA-R learn the weights of features by minimising an objective function which measures the average leave-one-out (LOO) classification or regression loss over the training data and fine-tuning a regularisation parameter ‘λ’ to 0.0329 while evaluating the weights of the features such that the significance is indicative of the rank.” Here, Szmulewicz teaches an objective function with a regularization parameter or term that evaluates weights or strength that change significance based on the rank or priority.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty and Watanabe with the teachings of Szmulewicz by using Chakravorty’s teachings of regularization techniques in regression model and Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, and incorporate with Szmulewicz’s teachings of an objective function with a regularization term.
One of ordinary skill in the art would be motivated to do so because by integrating Szmulewicz’s frameworks into the methods of Chakravorty and Watanabe, one of ordinary skill in the art would bring a “a second algorithmic model used to calculate a severity score which is indicative of severity of the motor control disorder in the subject, and the machine learning approach is a random-forest regression model (RFR).” (Szmulewicz, paragraph [0040]).
Claim 14:
Regarding claim 14, Chakravorty teaches “A computer program product comprising a computer-readable medium including programmed instructions, the instructions causing a computer to execute: detecting whether or not one or more conditions defining timings to perform learning of a regression model configured to predict one or more objective variables for a plurality of explanatory variables are satisfied;”
See Chakravorty in paragraph [0016] describing, “In addition, any of the methods, described above and elsewhere herein, may be embodied, individually or in any combination, in executable software modules of a processor-based system, such as a server, and/or in executable instructions stored in a non-transitory computer-readable medium.” Here, Chakravorty establishes the embodiment described being a method with a system or apparatus with a processor which is being interpreted as the processing unit, and a computer-readable medium which is being interpreted as the computer program product. Further, see Chakravorty in paragraph [0100] describing, “In subprocess 433, the parameters of machine-learning model 334 are determined and the error of the optimization problem is calculated. The parameters of machine-learning model 334 may comprise the weights and/or other variables in machine-learning model 334 (i.e., ƒθ). It should be understood that these weights are different than the weights λ assigned to the clusters in subprocess 432.” Here, Chakravorty establishes a subprocess that shows a model with various variables which are being interpreted as the explanatory variables. Further, see Chakravorty in paragraph [0101] describing, “In subprocess 434, it is determined whether or not a first stopping condition for the inner loop is satisfied. The first stopping condition may comprise a number of iterations (i.e., training epochs) of the inner loop reaching or exceeding a threshold, a change in the error of the optimization problem from the preceding iteration reaching or falling below a threshold, an error of the optimization problem reaching or falling below a threshold, a time duration of the inner loop reaching or exceeding a threshold, and/or the like. If the first stopping condition has been satisfied (i.e., “Yes” in subprocess 434), training subprocess 430 proceeds to subprocess 435.” Here, Chakravorty establishes detecting of a condition here for the learning of a model with the training. Further, see Chakravorty in paragraph [0118] describing, “Experiments were conducted using machine-learning model 334 with Laplace regularization for the task of fault localization in time domain protection. In particular, a regression model for estimating fault location was trained with Laplace regularization, using process 400, and without Laplace regularization.” Here, Chakravorty establishes the model for these processes as a regression model for estimating fault location which is the same as being configured to predict with the estimating, and the fault location can be the objective variable.
Further, Chakravorty teaches “and performing learning of the regression model by using an objective function and learning data,...”
See Chakravorty in paragraph [0124] describing, “Disclosed embodiments frame the training of a machine-learning model 334 as a multi-objective optimization problem. In addition to the primary objective function that calculates an error of machine-learning model 334 (e.g., as a regression task), the multi-objective optimization problem comprises a non-smoothness penalization function gθ, which drives the primary objective function to a solution that respects smoothness between the inputs and outputs of machine-learning model 334.” Here, Chakravorty establishes performing learning of the regression model with the training of the machine learning model established to be a regression model in previous limitation using a multi-objective optimization problem containing a primary objective function that comprises of a non-smoothness penalization function. Further, see Chakravorty in paragraph [0125] describing, “This Laplace regularization, using non-smoothness penalization function gθ, with or without adaptive weights for clusters, can be employed in any regression task in which the data are heterogeneous and the regression function is foreseen to be smooth on the data (i.e., the regression function does not vary abruptly with the data points).” Here, Chakravorty establishes Laplace regularization, which is known in the art to be a regularization term in itself, with the non-smoothness penalization function using data and the function with adaptive weights which can be seen as the regularization strength being changed according to priorities as the weight change according to priorities was established in previous limitations.
Chakravorty does not appear to explicitly teach “determine priorities of the plurality of explanatory variables according to a condition detected to be satisfied; …the objective function including a regularization term having a regularization strength changing according to the priorities.”,
However in the same field of art, Watanabe teaches “determine priorities of the plurality of explanatory variables according to a condition detected to be satisfied;”
See Watanabe in paragraph [0087] describing, “Specifically, the regression model generator 71 uses the generally known Random forest algorithm to calculate indexes of priority P for each of weather elements, demand factors, and data demand densities available as explanatory variables and normalizes the indexes to 0 through 1. Suppose the normalized value of priority P for each of the weather elements, the demand factors, and the data demand densities satisfies “the condition of the explanatory variable used for the regression model” notified from the reference point adjustment processor 70.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty with the teachings of Watanabe by using Chakravorty’s teachings of regularization techniques in regression model, and incorporate with Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models.
One of ordinary skill in the art would be motivated to do so because by integrating Watanabe’s frameworks into the methods of Chakravorty, which are both in the same field of art of regression models, one of ordinary skill in the art would bring a “an energy operation apparatus [that] operates energy in a management area based on a prediction result for supply and/or demand for the energy in the management area. The energy operation apparatus includes a demand predictor, a planner, an evaluator, and a solution quality controller. The demand predictor predicts demand and/or a power generation amount of future energy in a management area. The planner prepares a future energy supply plan in the management area based on a prediction result from the demand predictor. The evaluator evaluates supply and/or demand conditions including at least one of future weather in the management area, energy demand in the management area, a demand density and/or a generation density of future energy in the management area. The solution quality controller controls the quality of at least one of a prediction solution of the demand predictor and the energy supply plan of the planner based on an evaluation result from the evaluator.” (Watanabe, paragraph [0009]).
Neither Chakravorty or Watanabe appear to explicitly teach “…the objective function including a regularization term having a regularization strength changing according to the priorities.”,
However in the same field of art, Szmulewicz teaches “…the objective function including a regularization term having a regularization strength changing according to the priorities.”
See Szmulewicz in paragraph [0104] describing, “Random Subset Feature Selection (RSFS), Statistical Dependency (SDe) and Neighbourhood Component Analysis with regularization (NCA-R) were compared. NCA-R was selected because its low cross-validation error resulted in significantly fewer features. NCA-R learn the weights of features by minimising an objective function which measures the average leave-one-out (LOO) classification or regression loss over the training data and fine-tuning a regularisation parameter ‘λ’ to 0.0329 while evaluating the weights of the features such that the significance is indicative of the rank.” Here, Szmulewicz teaches an objective function with a regularization parameter or term that evaluates weights or strength that change significance based on the rank or priority.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty and Watanabe with the teachings of Szmulewicz by using Chakravorty’s teachings of regularization techniques in regression model and Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, and incorporate with Szmulewicz’s teachings of an objective function with a regularization term.
One of ordinary skill in the art would be motivated to do so because by integrating Szmulewicz’s frameworks into the methods of Chakravorty and Watanabe, one of ordinary skill in the art would bring a “a second algorithmic model used to calculate a severity score which is indicative of severity of the motor control disorder in the subject, and the machine learning approach is a random-forest regression model (RFR).” (Szmulewicz, paragraph [0040]).
Claim(s) 2-4 are rejected under 35 U.S.C. 103 as being unpatentable over Chakravorty J. et al, in view of Watanabe T. et al, further in view of Szmulewicz D. et al, and further in view of Miao J. et al, "A Dummy-Variable Model for Humidity-Influenced DC Film Capacitors Lifetime Estimation" available at https://ieeexplore.ieee.org/document/9895439?source=IQplus, effectively published on September 20, 2022, (hereafter Miao).
Claim 2:
Regarding claim 2, Chakravorty in view of Watanabe, and further in view of Szmulewicz teaches the limitations of claim 1.
Further, Chakravorty teaches “…and generate the learning data including the plurality of calculated explanatory variables;”
See Chakravorty in paragraph [0100] describing, “In subprocess 433, the parameters of machine-learning model 334 are determined and the error of the optimization problem is calculated. The parameters of machine-learning model 334 may comprise the weights and/or other variables in machine-learning model 334 (i.e., ƒθ). It should be understood that these weights are different than the weights λ assigned to the clusters in subprocess 432. In an initial iteration of subprocess 433, the parameters of machine-learning model 334 may be initialized to initial values. In each subsequent iteration of subprocess 433, the parameters may be updated. Again, the new parameters may be determined using any suitable technique, such as Pareto optimization, to identify a potential set of parameters that may further minimize the optimization problem.”
Further, Chakravorty teaches “and perform learning of the regression model using the generated learning data.”
See Chakravorty in paragraph [0100] describing, “In subprocess 433, the parameters of machine-learning model 334 are determined and the error of the optimization problem is calculated. The parameters of machine-learning model 334 may comprise the weights and/or other variables in machine-learning model 334 (i.e., ƒθ). It should be understood that these weights are different than the weights λ assigned to the clusters in subprocess 432. In an initial iteration of subprocess 433, the parameters of machine-learning model 334 may be initialized to initial values. In each subsequent iteration of subprocess 433, the parameters may be updated. Again, the new parameters may be determined using any suitable technique, such as Pareto optimization, to identify a potential set of parameters that may further minimize the optimization problem.” Further, see Chakravorty in paragraph [0118] describing, “Experiments were conducted using machine-learning model 334 with Laplace regularization for the task of fault localization in time domain protection. In particular, a regression model for estimating fault location was trained with Laplace regularization” Here, Chakravorty establishes the model explicitly as a regression model.
Chakravorty, does not appear to explicitly teach “The apparatus according to claim 1, wherein the regression model is configured to predict a plurality of objective variables, and the processing unit is configured to: calculate the plurality of explanatory variables by multiplying one or more first variables and a plurality of dummy variables corresponding to the plurality of objective variables,”,
Further, Watanabe teaches “The apparatus according to claim 1, wherein the regression model is configured to predict a plurality of objective variables,…”
See Watanabe in paragraph [0090] describing, “Suppose the solution quality controller 36 supplies predictive solution target stabilization index “1.” This generates an exact “Ridge regression model” having many explanatory variables represented by the weather element, the demand factor, and the data demand density whose normalized priority P is “0.2” or more. This can precisely predict an energy demand. In particular, a highly precise predicted value is available when the weather or a demand factor fits in a normal range and the prediction using the regression model is easily performed.” Here, Watanabe establishes a prediction of a plurality of objective variables, the objective here being the energy demand, with the many explanatory variables used for prediction and the prediction is done using a regression model.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty with the teachings of Watanabe by using Chakravorty’s teachings of regularization techniques in regression model, and incorporate with Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models.
One of ordinary skill in the art would be motivated to do so because by integrating Watanabe’s frameworks into the methods of Chakravorty, which are both in the same field of art of regression models, one of ordinary skill in the art would bring a “an energy operation apparatus [that] operates energy in a management area based on a prediction result for supply and/or demand for the energy in the management area. The energy operation apparatus includes a demand predictor, a planner, an evaluator, and a solution quality controller. The demand predictor predicts demand and/or a power generation amount of future energy in a management area. The planner prepares a future energy supply plan in the management area based on a prediction result from the demand predictor. The evaluator evaluates supply and/or demand conditions including at least one of future weather in the management area, energy demand in the management area, a demand density and/or a generation density of future energy in the management area. The solution quality controller controls the quality of at least one of a prediction solution of the demand predictor and the energy supply plan of the planner based on an evaluation result from the evaluator.” (Watanabe, paragraph [0009]).
None of Chakravorty, Watanabe, or Szmulewicz appear to explicitly teach “…and the processing unit is configured to: calculate the plurality of explanatory variables by multiplying one or more first variables and a plurality of dummy variables corresponding to the plurality of objective variables,”,
However in the same field of art, Miao teaches “…and the processing unit is configured to: calculate the plurality of explanatory variables by multiplying one or more first variables and a plurality of dummy variables corresponding to the plurality of objective variables,”
See Miao in Introduction section on page 2 describing, “The multivariate linear regression (MLR) algorithm is a suitable method for small data with various categories. In the MLR model, categorical variables can be transformed into dummy variables to build estimation models properly [32], [33], [34], [35], [36], [37]. Various methods that can handle dummy variables are described and compared in [32], which concludes that the most effective method is different for datasets and the stratification and dummy variables-based method should be tried at least. Besides, according to [32], although the stratification is flexible which divides the dataset into subsets and builds models for each subset, its estimation accuracy may be low because of the small subset, especially when there are many independent variables. In comparison, models with dummy variables may achieve higher estimation accuracy but require fewer data. If a dataset has p noncategorical variables and q categorical variables, it would need 5(p+q−1 ) data points for dummy variables while 5pq data points for the stratification [32]. Moreover, interaction terms made by multiplying independent variables and dummy variables also need to be introduced to the MLR algorithm if necessary [32].” Here, Miao establishes a regression model, that calculates independent variables, which as known in the art is an explanatory variable, with dummy variables by multiplying both the independent variables which can be first variables and dummy variables which correspond to categorical variables which can be seen as objective variables.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty, Watanabe, and Szmulewicz with the teachings of Miao by using Chakravorty’s teachings of regularization techniques in regression model, Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, and Szmulewicz’s teachings of an objective function, and incorporate with Miao’s teachings of a dummy-variable model.
One of ordinary skill in the art would be motivated to do so because by integrating Miao’s frameworks into the methods of Chakravorty, Watanabe, and Szmulewicz one of ordinary skill in the art would bring a “proposed method [that] is data-driven without establishing a complex mathematical model based on the physics-of-failure mechanism. Compared to Peck’s model, it can track capacitor degradation with time. Since this method uses small samples to infer the population distribution, it does not require a large amount of data.”, “proposed method introduces the humidity accelerating factor into the lifetime estimation model utilizing dummy variables. It mixes all the humidity data and only does one regression, thus increasing the degrees of freedom (DFs) and improving the estimation accuracy of the regression coefficients.”, “The tradeoff between the amount of data and estimation accuracy is discussed and it indicates that the proposed lifetime estimation model can achieve comparable accuracy with reduced data and testing time.”, and “The proposed model has wide applicability. Since it is data-driven, the aging data with the same characteristics can also be evaluated with it. The degradation indicator in this study follows an exponential distribution, which is very common in humidity-influenced aging of capacitors and other electronic components.” (Miao, Introduction section on page 2).
Claim 3:
Regarding claim 3, Chakravorty in view of Watanabe, further in view of Szmulewicz, and further in view of Miao teaches the limitations of claim 2.
Chakravorty does not appear to explicitly teach “The apparatus according to claim 2, wherein the processing unit is configured to: specify a type of objective variable corresponding to a dummy variable as a type of explanatory variable, and generate first correspondence information in which the specified type of explanatory variable and the explanatory variable are associated with each other; and determine a priority corresponding to a type of the detected condition for the explanatory variable included in the type of explanatory variable corresponding to the type of the detected condition by using the first correspondence information and second correspondence information in which the type of condition, the type of explanatory variable, and the priority are associated with each other.”,
Further, Watanabe teaches “and determine a priority corresponding to a type of the detected condition for the explanatory variable included in the type of explanatory variable corresponding to the type of the detected condition by using…second correspondence information in which the type of condition, the type of explanatory variable, and the priority are associated with each other.”
See Watanabe in paragraph [0084] describing, “Actually, the reference point adjustment processor 70 maintains a second table 73 as illustrated in FIG. 8. The second table 73 defines types of the regression model to be generated by the regression model generator 71 and conditions of an explanatory variable (each weather element, each demand factor, and/or data demand density) used for the regression model corresponding to values of the predictive solution target stabilization index specified by the solution quality controller 36. A user previously creates the second table 73. In this case, the condition of the explanatory variable used for the regression model is provided as a threshold value (the “threshold” column in FIG. 8) of priority P for the explanatory variable.” Further, see Watanabe in paragraph [0085] describing, “The condition of the explanatory variable corresponds to the priority value of the explanatory variable. The reference point adjustment processor 70 selects and outputs the regression model and the condition of the explanatory variable used for the regression model to the regression model generator 71.” Further, see Watanabe in paragraph [0087] describing, “Specifically, the regression model generator 71 uses the generally known Random forest algorithm to calculate indexes of priority P for each of weather elements, demand factors, and data demand densities available as explanatory variables and normalizes the indexes to 0 through 1. Suppose the normalized value of priority P for each of the weather elements, the demand factors, and the data demand densities satisfies “the condition of the explanatory variable used for the regression model” notified from the reference point adjustment processor 70. Then, the regression model generator 71 uses these weather elements, demand factors, and data demand densities as explanatory variables to generate the regression model notified from the reference point adjustment processor 70.”Here, Watanabe establishes a priority of a condition that is determined with a threshold value to that corresponds to the explanatory variable. The type of condition and explanatory variable corresponds to each other and examples of the types are given with weather elements for example. The second table established with this information can be seen as the second correspondence information.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Chakravorty with the teachings of Watanabe by using Chakravorty’s teachings of regularization techniques in regression model, and incorporate with Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models.
One of ordinary skill in the art would be motivated to do so because by integrating Watanabe’s frameworks into the methods of Chakravorty, which are both in the same field of art of regression models, one of ordinary skill in the art would bring a “an energy operation apparatus [that] operates energy in a management area based on a prediction result for supply and/or demand for the energy in the management area. The energy operation apparatus includes a demand predictor, a planner, an evaluator, and a solution quality controller. The demand predictor predicts demand and/or a power generation amount of future energy in a management area. The planner prepares a future energy supply plan in the management area based on a prediction result from the demand predictor. The evaluator evaluates supply and/or demand conditions including at least one of future weather in the management area, energy demand in the management area, a demand density and/or a generation density of future energy in the management area. The solution quality controller controls the quality of at least one of a prediction solution of the demand predictor and the energy supply plan of the planner based on an evaluation result from the evaluator.” (Watanabe, paragraph [0009]).
None of Chakravorty, Watanabe, or Szmulewicz appear to explicitly teach “The apparatus according to claim 2, wherein the processing unit is configured to: specify a type of objective variable corresponding to a dummy variable as a type of explanatory variable, and generate first correspondence information in which the specified type of explanatory variable and the explanatory variable are associated with each other; …the first correspondence information...”,
However in the same field of art, Miao teaches “The apparatus according to claim 2, wherein the processing unit is configured to: specify a type of objective variable corresponding to a dummy variable as a type of explanatory variable,”
See Miao in Introduction section on page 2 describing, “The multivariate linear regression (MLR) algorithm is a suitable method for small data with various categories. In the MLR model, categorical variables can be transformed into dummy variables to build estimation models properly [32], [33], [34], [35], [36], [37]. Various methods that can handle dummy variables are described and compared in [32], which concludes that the most effective method is different for datasets and the stratification and dummy variables-based method should be tried at least. Besides, according to [32], although the stratification is flexible which divides the dataset into subsets and builds models for each subset, its estimation accuracy may be low because of the small subset, especially when there are many independent variables. In comparison, models with dummy variables may achieve higher estimation accuracy but require fewer data. If a dataset has p noncategorical variables and q categorical variables, it would need 5(p+q−1 ) data points for dummy variables while 5pq data points for the stratification [32]. Moreover, interaction terms made by multiplying independent variables and dummy variables also need to be introduced to the MLR algorithm if necessary [32].” Here, Miao establishes a regression model, that calculates independent variables, which as known in the art is an explanatory variable, with dummy variables by multiplying both the independent variables which can be first variables and dummy variables which correspond to categorical variables which can be seen as objective variables.
Further, Miao teaches “and generate first correspondence information in which the specified type of explanatory variable and the explanatory variable are associated with each other;”
See Miao Table I and section II. ACCELERATED AGING TEST FOR DC FILM CAPACITORS UNDER HUMIDITY CONDITIONS on page 3 describing, “
PNG
media_image1.png
160
330
media_image1.png
Greyscale
For each group, ten MPPF capacitors are tested. During the degradation process, the testing proceeds until most or all samples fail. Therefore, different humidity stresses determine unequal testing time. In this study, the testing under 55%, 70%, and 85% RH last for 3850, 2700, and 2160 h, respectively. As a result, the dataset consists of a total of 30 time-series data from the aging tests of three groups. The dataset details are summarized in Table I.” Here, Miao establishes generating correspondence information with the dataset from testing in table I and there is a specified type of explanatory variable here with the relative humidity and temperature condition.
Further, Miao teaches “…using the first correspondence information...”
See Miao Table I and section II. ACCELERATED AGING TEST FOR DC FILM CAPACITORS UNDER HUMIDITY CONDITIONS on page 3 describing, “
PNG
media_image1.png
160
330
media_image1.png
Greyscale
For each group, ten MPPF capacitors are tested. During the degradation process, the testing proceeds until most or all samples fail. Therefore, different humidity stresses determine unequal testing time. In this study, the testing under 55%, 70%, and 85% RH last for 3850, 2700, and 2160 h, respectively. As a result, the dataset consists of a total of 30 time-series data from the aging tests of three groups. The dataset details are summarized in Table I.” Here, Miao establishes first correspondence information with the dataset from testing in table I and there is a specified type of explanatory variable here with the relative humidity and temperature condition which can be seen as the first correspondence information.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty, Watanabe, and Szmulewicz with the teachings of Miao by using Chakravorty’s teachings of regularization techniques in regression model, Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, and Szmulewicz’s teachings of an objective function, and incorporate with Miao’s teachings of a dummy-variable model.
One of ordinary skill in the art would be motivated to do so because by integrating Miao’s frameworks into the methods of Chakravorty, Watanabe, and Szmulewicz one of ordinary skill in the art would bring a “proposed method [that] is data-driven without establishing a complex mathematical model based on the physics-of-failure mechanism. Compared to Peck’s model, it can track capacitor degradation with time. Since this method uses small samples to infer the population distribution, it does not require a large amount of data.”, “proposed method introduces the humidity accelerating factor into the lifetime estimation model utilizing dummy variables. It mixes all the humidity data and only does one regression, thus increasing the degrees of freedom (DFs) and improving the estimation accuracy of the regression coefficients.”, “The tradeoff between the amount of data and estimation accuracy is discussed and it indicates that the proposed lifetime estimation model can achieve comparable accuracy with reduced data and testing time.”, and “The proposed model has wide applicability. Since it is data-driven, the aging data with the same characteristics can also be evaluated with it. The degradation indicator in this study follows an exponential distribution, which is very common in humidity-influenced aging of capacitors and other electronic components.” (Miao, Introduction section on page 2).
Claim 4:
Regarding claim 4, Chakravorty in view of Watanabe, further in view of Szmulewicz, further in view of Miao teaches the limitations of claim 2.
Further, Chakravorty teaches ”The apparatus according to claim 2, wherein the processing unit is configured to determine a priority of an explanatory variable corresponding to an objective variable according to a magnitude of a prediction error by the regression model after learning among the plurality of objective variables.”
See Chakravorty in paragraph [0100] describing, “In subprocess 433, the parameters of machine-learning model 334 are determined and the error of the optimization problem is calculated. The parameters of machine-learning model 334 may comprise the weights and/or other variables in machine-learning model 334 (i.e., ƒθ). It should be understood that these weights are different than the weights λ assigned to the clusters in subprocess 432. In an initial iteration of subprocess 433, the parameters of machine-learning model 334 may be initialized to initial values. In each subsequent iteration of subprocess 433, the parameters may be updated.” Here, Chakravorty establishes the parameters consisting of weights and variables that are adjusted and updated which can be seen as determining priority with the weight adjustment and the variables can be seen as explanatory variables. Further, see Chakravorty in paragraph [0101] describing, “In subprocess 434, it is determined whether or not a first stopping condition for the inner loop is satisfied. The first stopping condition may comprise a number of iterations (i.e., training epochs) of the inner loop reaching or exceeding a threshold, a change in the error of the optimization problem from the preceding iteration reaching or falling below a threshold, an error of the optimization problem reaching or falling below a threshold, a time duration of the inner loop reaching or exceeding a threshold, and/or the like. If the first stopping condition has been satisfied (i.e., “Yes” in subprocess 434), training subprocess 430 proceeds to subprocess 435. In this case, the current iteration of the inner loop has ended. If the first stopping condition has not yet been satisfied (i.e., “No” in subprocess 434), training subprocess 430 returns to subprocess 433 to update the parameters of machine-learning model 334.” Here Chakravorty establishes the parameters being updated based on a change in error exceeding a threshold of an optimization problem, which is being interpreted in this specific embodiment as the magnitude of a prediction error, to then determine and update the parameters after learning or training.
Claim(s) 6 is rejected under 35 U.S.C. 103 as being unpatentable over Chakravorty J. et al, in view of Watanabe T. et al, further in view of Szmulewicz D. et al, and further in view of Sadanaga Y. et al, (Japanese Patent 2022184205 A) effectively filed on May 31, 2021, (hereafter Sadanaga).
Claim 6:
Regarding claim 6, Chakravorty in view of Watanabe, and further in view of Szmulewicz teaches the limitations of claim 1.
Further, Chakravorty teaches “and perform learning of the regression model using the generated learning data.”
See Chakravorty in paragraph [0100] describing, “In subprocess 433, the parameters of machine-learning model 334 are determined and the error of the optimization problem is calculated. The parameters of machine-learning model 334 may comprise the weights and/or other variables in machine-learning model 334 (i.e., ƒθ). It should be understood that these weights are different than the weights λ assigned to the clusters in subprocess 432. In an initial iteration of subprocess 433, the parameters of machine-learning model 334 may be initialized to initial values. In each subsequent iteration of subprocess 433, the parameters may be updated. Again, the new parameters may be determined using any suitable technique, such as Pareto optimization, to identify a potential set of parameters that may further minimize the optimization problem.” Further, see Chakravorty in paragraph [0118] describing, “Experiments were conducted using machine-learning model 334 with Laplace regularization for the task of fault localization in time domain protection. In particular, a regression model for estimating fault location was trained with Laplace regularization” Here, Chakravorty establishes the model explicitly as a regression model.
None of Chakravorty, Watanabe, or Szmulewicz appear to explicitly teach “The apparatus according to claim 1, wherein the processing unit is configured to: generate the learning data including one or more explanatory variables having priorities higher than another explanatory variable among the plurality of explanatory variables, and one or more objective variables;”,
However in the same field of art, Sadanaga teaches “The apparatus according to claim 1, wherein the processing unit is configured to: generate the learning data including one or more explanatory variables having priorities higher than another explanatory variable among the plurality of explanatory variables, and one or more objective variables;”
See Sadanaga on page 13 describing, “In each of these stratified datasets Ds1 to D4, the generation unit 128 uses one or more records included in the stratified dataset to generate data for each of the explanatory variables X0 and X1 and the objective variable Y Generate a model that shows the relationship between Thus, in the present embodiment, stratified variables are unused variables, stratified classification is performed according to the unused variables, and models are generated for each of a plurality of layers. Therefore, it is possible to optimally classify the data set Ds by stratification using the unused variables, which are variables other than the explanatory variables. As a result, even if the correlation between the explanatory variable and the objective variable changes according to the unused variable, it is possible to generate a highly accurate model according to the group of the unused variable.” Here, Sadanaga establishes generating learning data for each of the explanatory variables and objective variable using stratified datasets. Further, see Sadanaga on page 13 describing, “Then, the first variable identifying unit 121identifies explanatory variables and objective variables from the plurality of variables indicated by the data set Ds (step S2). As a result, explanatory variables and objective variables are set. For example, as described above, the variables X0 and X1 are set as explanatory variables, respectively, and the variable Y is set as the objective variable. Next, the stratification condition setting unit 122 sets the total number N of stratification variables according to the user's input operation (step S3). For example, the total number N is set to N=2. Then, the stratification condition setting unit 122 determines the priority order of the variable types of the stratification variables (step S4). That is, the priority of qualitative and quantitative variables is determined. For example, the priority of these variables is determined in order of qualitative variables and quantitative variables. Next, the unused variable extraction unit 123 extracts variables other than explanatory variables and objective variables as unused variables from the plurality of variables of the data set Ds (step S5).” Here, Sadanaga establishes priorities of the variables in the data and variables that are unused are the ones without priority.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty, Watanabe, and Szmulewicz with the teachings of Sadanaga by using Chakravorty’s teachings of regularization techniques in regression model, Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, and Szmulewicz’s teachings of an objective function with a regularization term, and incorporate with Sadanaga’s teachings of a model generation method and apparatus.
One of ordinary skill in the art would be motivated to do so because by integrating Sadanaga’s frameworks into the methods of Chakravorty, Watanabe, and Szmulewicz, one of ordinary skill in the art would bring “a model generation apparatus capable of easily improving accuracy of a model.” (Sadanaga, Abstract on page 1).
Claim(s) 7 is rejected under 35 U.S.C. 103 as being unpatentable over Chakravorty in view of Watanabe, and further in view of Szmulewicz, in view of Kobayashi K. et al, (US. Patent Application Publication 20190197435 A1) effectively filed on November 27, 2018, (hereafter Kobayashi).
Claim 7:
Regarding claim 7, Chakravorty in view of Watanabe, and further in view of Szmulewicz teaches the limitations of claim 1.
Chakravorty does not appear to explicitly teach “The apparatus according to claim 1, wherein the processing unit is configured to obtain changes of the plurality of explanatory variables between the learning data, and test data serving as an input in prediction using the regression model after learning, and determine the priorities of the plurality of explanatory variables according to magnitudes of the changes.”,
Further, Watanabe teaches “…and determine the priorities of the plurality of explanatory variables according to magnitudes of the changes.”
See Watanabe in paragraph [0087] describing, “Specifically, the regression model generator 71 uses the generally known Random forest algorithm to calculate indexes of priority P for each of weather elements, demand factors, and data demand densities available as explanatory variables and normalizes the indexes to 0 through 1. Suppose the normalized value of priority P for each of the weather elements, the demand factors, and the data demand densities satisfies “the condition of the explanatory variable used for the regression model” notified from the reference point adjustment processor 70. Then, the regression model generator 71 uses these weather elements, demand factors, and data demand densities as explanatory variables to generate the regression model notified from the reference point adjustment processor 70.”Here, Watanabe establishes determining priorities of explanatory variables. Further, see Watanabe in paragraph [0092] describing, “The regressive prediction processor 72 is a functional part that is implemented when the CPU 11 executes a regressive prediction processing program 72P (FIG. 5) stored in the external storage unit 13. The regressive prediction processor 72 acquires a value of the necessary explanatory variable from the weather data storer 41, the demand factor data storer 42, and a simulation result from the physical simulator 63 of the second predictor 32. The regressive prediction processor 72 calculates a predicted value for the energy demand at the time slice (specified time) in a management area for the relevant energy operation apparatus 10 so as to apply the acquired value to the regression model supplied from the regression model generator 71. The regressive prediction processor 72 outputs the calculated predicted value for the energy demand to a planning processor 101 of the planner 37 to be described later.” Further, see Watanabe in paragraph [0095] describing, “The temporal data distribution profile evaluator 80 uses, for example, a dispersion state (dispersion value) of the predicted values to evaluate the magnitude of the temporal change in the prediction results and calculate a weighted average.” Here, Watanabe establishes a magnitude of temporal change based on predicted values. These predicted values come from the regressive prediction processor using values from explanatory variables, with the explanatory variables relying on prediction the priorities set for the variables can and will be determined based on changes.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Chakravorty with the teachings of Watanabe by using Chakravorty’s teachings of regularization techniques in regression model, and incorporate with Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models.
One of ordinary skill in the art would be motivated to do so because by integrating Watanabe’s frameworks into the methods of Chakravorty, which are both in the same field of art of regression models, one of ordinary skill in the art would bring a “an energy operation apparatus [that] operates energy in a management area based on a prediction result for supply and/or demand for the energy in the management area. The energy operation apparatus includes a demand predictor, a planner, an evaluator, and a solution quality controller. The demand predictor predicts demand and/or a power generation amount of future energy in a management area. The planner prepares a future energy supply plan in the management area based on a prediction result from the demand predictor. The evaluator evaluates supply and/or demand conditions including at least one of future weather in the management area, energy demand in the management area, a demand density and/or a generation density of future energy in the management area. The solution quality controller controls the quality of at least one of a prediction solution of the demand predictor and the energy supply plan of the planner based on an evaluation result from the evaluator.” (Watanabe, paragraph [0009]).
None of Chakravorty, Watanabe, or Szmulewicz appear to explicitly teach “The apparatus according to claim 1, wherein the processing unit is configured to obtain changes of the plurality of explanatory variables between the learning data, and test data serving as an input in prediction using the regression model after learning,...”,
However in the same field of art, Kobayashi teaches “The apparatus according to claim 1, wherein the processing unit is configured to obtain changes of the plurality of explanatory variables between the learning data, and test data serving as an input in prediction using the regression model after learning,”
See Kobayashi in paragraph [0065] describing, “The machine learning apparatus 100 calculates the “prediction performance” of a learned model. The prediction performance is the capability of accurately predicting results of unknown cases and may be referred to as “accuracy”. The machine learning apparatus 100 samples unit data other than the training data from the collected data as test data and calculates the prediction performance by using the test data. The size of the test data is about half the size of the training data, for example. The machine learning apparatus 100 inputs the values of the explanatory variables included in the test data to the model and compares the values (predicted values) of the objective variables that the model outputs with the values (result values) of the objective variables included in the test data. Hereinafter, evaluating the prediction performance of a learned model will be referred to as “validation”, as needed.” Further, see Kobayashi in paragraph [0040] describing, “To generate the model, various kinds of machine learning algorithms may be used such as a logistic regression analysis, a support vector machine, and a random forest.” Here, Kobayashi establishes changes of the explanatory variables between learning and test data with the input of the variables in the test data and comparing to predicted values which is being interpreted as learning data. The input is to a model, which can be seen as a regression model with the model and its algorithms doing logistic regression, that calculates prediction.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty, Watanabe, and Szmulewicz with the teachings of Kobayashi by using Chakravorty’s teachings of regularization techniques in regression model, Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, and Szmulewicz’s teachings of an objective function with a regularization term, and incorporate with Kobayashi’s teachings of a estimate method and apparatus.
One of ordinary skill in the art would be motivated to do so because by integrating Kobayashi’s frameworks into the methods of Chakravorty, Watanabe, and Szmulewicz, one of ordinary skill in the art would bring a “estimation apparatus 10 according to the first embodiment estimates a prediction performance curve which indicates a relationship between data sizes of training data used for machine learning and prediction performances of a model generated by the machine learning.” (Kobayashi, paragraph [0038]).
Claim(s) 8 is rejected under 35 U.S.C. 103 as being unpatentable over Chakravorty J. et al, in view of Watanabe T. et al, further in view of Szmulewicz D. et al, further in view of Sadanaga Y. et al, and further in view of Takada M. et al, " Transfer Learning via ℓ1 Regularization" available at https://proceedings.neurips.cc/paper_files/paper/2020/file/a4a83056b58ff983d12c72bb17996243-Paper.pdf, effectively published in 2020, (hereafter Takada).
Claim 8:
Regarding claim 8, Chakravorty in view of Watanabe and further in view of Szmulewicz teaches the limitations of claim 1.
Further, Chakravorty teaches “…and the objective function includes: a term evaluating compatibility between a prediction result and correct data;”
See Chakravorty in paragraph [0075] describing, “In subprocess 430, machine-learning model 334 is trained by minimizing the error in the optimization problem, generated in subprocess 420. It should be understood that the error is represented by the objective function which calculates a difference between the ground truth and the output of machine-learning model 334.” Here, Chakravorty establishes an objective function using error, which is being interpreted as the term, that evaluates or calculates a difference or compatibility of ground truth data which is correct data, with an output of the model which can be seen as a prediction result.
None of Chakravorty, Watanabe, or Szmulewicz appear to explicitly teach “The apparatus according to claim 1, wherein the priorities include selection priorities representing priorities of selecting the plurality of explanatory variables and update priorities representing priorities of updating the plurality of explanatory variables,…a first regularization term obtained by multiplying terms for regularizing parameters corresponding to the plurality of explanatory variables by weights based on the selection priorities; and a second regularization term obtained by multiplying terms for regularizing changes with respect to the parameters before update by weights based on the update priorities.”,
However in the same field of art, Sadanaga teaches “The apparatus according to claim 1, wherein the priorities include selection priorities representing priorities of selecting the plurality of explanatory variables and update priorities representing priorities of updating the plurality of explanatory variables,…”
See Sadanaga on page 1 describing, “the model generator selects objective variables and explanatory variables from a data set containing data for each of a plurality of variables, and generates a correlation coefficient between those variables or a regression model using those variables.” Here, Sadanaga establishes selecting and updating of explanatory variables using the model generator. Further, see Sadanaga on page 13 describing, “First, the reception unit 130 of the model generation device 100 performs data reception processing (step S1). In this data reception process, the first variable identification unit 121 receives the data set Ds by reading the data set Ds from the database 106 . Then, the first variable identifying unit 121 identifies explanatory variables and objective variables from the plurality of variables indicated by the data set Ds (step S2). As a result, explanatory variables and objective variables are set. For example, as described above, the variables X0 and X1 are set as explanatory variables, respectively, and the variable Y is set as the objective variable. Next, the stratification condition setting unit 122 sets the total number N of stratification variables according to the user's input operation (step S3). For example, the total number N is set to N=2. Then, the stratification condition setting unit 122 determines the priority order of the variable types of the stratification variables (step S4). That is, the priority of qualitative and quantitative variables is determined. For example, the priority of these variables is determined in order of qualitative variables and quantitative variables.” Here, Sadanaga establishes the selecting and updating of explanatory variables using the model generator with determining priorities.
Further, Sadanaga teaches “…the plurality of explanatory variables by weights based on the selection priorities…”
See Sadanaga on page 1 describing, “the model generator selects objective variables and explanatory variables from a data set containing data for each of a plurality of variables, and generates a correlation coefficient between those variables or a regression model using those variables.” Here, Sadanaga establishes selecting and updating of explanatory variables using the model generator. Further, see Sadanaga on page 13 describing, “First, the reception unit 130 of the model generation device 100 performs data reception processing (step S1). In this data reception process, the first variable identification unit 121 receives the data set Ds by reading the data set Ds from the database 106 . Then, the first variable identifying unit 121 identifies explanatory variables and objective variables from the plurality of variables indicated by the data set Ds (step S2). As a result, explanatory variables and objective variables are set. For example, as described above, the variables X0 and X1 are set as explanatory variables, respectively, and the variable Y is set as the objective variable. Next, the stratification condition setting unit 122 sets the total number N of stratification variables according to the user's input operation (step S3). For example, the total number N is set to N=2. Then, the stratification condition setting unit 122 determines the priority order of the variable types of the stratification variables (step S4). That is, the priority of qualitative and quantitative variables is determined. For example, the priority of these variables is determined in order of qualitative variables and quantitative variables.” Here, Sadanaga establishes the selecting and updating of explanatory variables using the model generator with determining priorities.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty, Watanabe, and Szmulewicz with the teachings of Sadanaga by using Chakravorty’s teachings of regularization techniques in regression model, Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, and Szmulewicz’s teachings of an objective function with a regularization term and incorporate with Sadanaga’s teachings of a model generation method and apparatus.
One of ordinary skill in the art would be motivated to do so because by integrating Sadanaga’s frameworks into the methods of Chakravorty, Watanabe, and Szmulewicz, one of ordinary skill in the art would bring “a model generation apparatus capable of easily improving accuracy of a model.” (Sadanaga, Abstract on page 1).
None of Chakravorty, Watanabe, Szmulewicz, or Sadanaga appears to explicitly teach “a first regularization term obtained by multiplying terms for regularizing parameters corresponding to the plurality of explanatory variables by weights based on the selection priorities; and a second regularization term obtained by multiplying terms for regularizing changes with respect to the parameters before update by weights based on the update priorities.”,
However, Takada teaches “a first regularization term obtained by multiplying terms for regularizing parameters corresponding to the plurality of explanatory variables by weights based on the selection priorities;”
See Takada section 2 Methods on pages 2-3 describing, “Suppose that we have an initial estimate of β∗ as ˜β ∈ Rp and that the initial estimate is associated with the present estimate. Then, a natural assumption is that the difference between initial and present estimates is sparse. Thus, we employ the ℓ1 regularization of the estimate difference and incorporate it into the ordinary Lasso regularization a
PNG
media_image2.png
62
510
media_image2.png
Greyscale
where λ = λn > 0 and 0 ≤ α ≤ 1 are regularization parameters.” Here, Takada gives an equation and describes the first and second regularization terms obtained by multiplying terms. The first is for regularizing parameters, the second is regularizing changes. α and λ are weights based on the selection priorities, 1 – α and λ are weights based on the update priorities. β are parameters corresponding to the plurality of explanatory variables, β – β’ is changes with respect to the parameters.
Further, Takada teaches “and a second regularization term obtained by multiplying terms for regularizing changes with respect to the parameters before update by weights based on the update priorities.”
See Takada section 2 Methods on pages 2-3 describing, “Suppose that we have an initial estimate of β∗ as ˜β ∈ Rp and that the initial estimate is associated with the present estimate. Then, a natural assumption is that the difference between initial and present estimates is sparse. Thus, we employ the ℓ1 regularization of the estimate difference and incorporate it into the ordinary Lasso regularization a
PNG
media_image2.png
62
510
media_image2.png
Greyscale
where λ = λn > 0 and 0 ≤ α ≤ 1 are regularization parameters.” Here, Takada gives an equation and describes the first and second regularization terms obtained by multiplying terms. The second is regularizing changes. α and λ are weights based on the selection priorities, (1 – α) and λ are weights based on the update priorities. β are parameters corresponding to the plurality of explanatory variables, β – β’ is changes with respect to the parameters.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty, Watanabe, Szmulewicz, and Sadanaga with the teachings of Takada by using Chakravorty’s teachings of regularization techniques in regression model, Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, Szmulewicz’s teachings of an objective function with a regularization term, and Sadanaga’s teachings of a model generation method and apparatus and incorporate with Takada’s teaching of a regularization function.
One of ordinary skill in the art would be motivated to do so because by integrating Takada’s frameworks into the methods of Chakravorty, Watanabe, Szmulewicz, and Sadanaga, one of ordinary skill in the art would bring a “method with clear theoretical justifications. We demonstrate three favorable characteristics for our method. First our method presents a smaller estimation error than Lasso when the underlying functions do not change and the source estimate is the same as a target parameter. This indicates that our method effectively transfers knowledge under a stationary environment. Second, our method gives a consistent estimate even when the source estimate is mistaken, albeit with a weak convergence rate due to the phenomenon of so-called negative transfers[42] bracket. This implies that our method can effectively discard the outdated knowledge and obtain new knowledge under nonstationary environment. Third, our method does not update estimates when the residuals of the predictions are small and the regularization is large. Hence, our method has an implicit station narrative detection mechanism.” (Takada, page 2 section 1 Introduction).
Claim(s) 12 is rejected under 35 U.S.C. 103 as being unpatentable over Chakravorty J. et al, in view of Watanabe T. et al, further in view of Szmulewicz D. et al, and further in view of Sawada A. et al, (US. Patent Application Publication 20210133474 A1) effectively filed on May 18, 2018.
Claim 12:
Regarding claim 12, Chakravorty in view of Watanabe, and further in view of Szmulewicz teaches the limitations of claim 1.
Further, Chakravorty teaches “The apparatus according to claim 1, wherein the processing unit comprises: a detection unit configured to detect whether or not the one or more conditions are satisfied;”
See Chakravorty in Fig. 3 showing “
PNG
media_image3.png
290
458
media_image3.png
Greyscale
”. Here, Chakravorty shows a controller in a line protection system that controls the ML model. With the broadest reasonable interpretation, the controller can act as the detection unit for the ML model. Further, see Chakravorty in paragraph [0100] describing, “In subprocess 433, the parameters of machine-learning model 334 are determined and the error of the optimization problem is calculated. The parameters of machine-learning model 334 may comprise the weights and/or other variables in machine-learning model 334 (i.e., ƒθ). It should be understood that these weights are different than the weights λ assigned to the clusters in subprocess 432. In an initial iteration of subprocess 433, the parameters of machine-learning model 334 may be initialized to initial values. In each subsequent iteration of subprocess 433, the parameters may be updated. Again, the new parameters may be determined using any suitable technique, such as Pareto optimization, to identify a potential set of parameters that may further minimize the optimization problem.” Further, see Chakravorty in paragraph [0101] describing, “In subprocess 434, it is determined whether or not a first stopping condition for the inner loop is satisfied. The first stopping condition may comprise a number of iterations (i.e., training epochs) of the inner loop reaching or exceeding a threshold, a change in the error of the optimization problem from the preceding iteration reaching or falling below a threshold, an error of the optimization problem reaching or falling below a threshold, a time duration of the inner loop reaching or exceeding a threshold, and/or the like. If the first stopping condition has been satisfied (i.e., “Yes” in subprocess 434), training subprocess 430 proceeds to subprocess 435.” Here, Chakravorty establishes detecting of a condition here for the learning of a model with the training and determining weights which as established can be seen to be determining priorities.
Further, Chakravorty teaches “and a learning unit configured to perform learning of the regression model.”
See Chakravorty in Fig. 3 showing “
PNG
media_image3.png
290
458
media_image3.png
Greyscale
”. Here, Chakravorty shows a controller in a line protection system that controls the ML model. With the broadest reasonable interpretation, the controller can act as each the detection, and learning unit for the ML model. Further, see Chakravorty in paragraph [0100] describing, “In subprocess 433, the parameters of machine-learning model 334 are determined and the error of the optimization problem is calculated. The parameters of machine-learning model 334 may comprise the weights and/or other variables in machine-learning model 334 (i.e., ƒθ). It should be understood that these weights are different than the weights λ assigned to the clusters in subprocess 432. In an initial iteration of subprocess 433, the parameters of machine-learning model 334 may be initialized to initial values. In each subsequent iteration of subprocess 433, the parameters may be updated. Again, the new parameters may be determined using any suitable technique, such as Pareto optimization, to identify a potential set of parameters that may further minimize the optimization problem.” Further, see Chakravorty in paragraph [0101] describing, “In subprocess 434, it is determined whether or not a first stopping condition for the inner loop is satisfied. The first stopping condition may comprise a number of iterations (i.e., training epochs) of the inner loop reaching or exceeding a threshold, a change in the error of the optimization problem from the preceding iteration reaching or falling below a threshold, an error of the optimization problem reaching or falling below a threshold, a time duration of the inner loop reaching or exceeding a threshold, and/or the like. If the first stopping condition has been satisfied (i.e., “Yes” in subprocess 434), training subprocess 430 proceeds to subprocess 435.” Here, Chakravorty establishes the learning of a model with the training and determining weights which as established can be seen to be determining priorities. Further, see Chakravorty in paragraph [0118] describing, “Experiments were conducted using machine-learning model 334 with Laplace regularization for the task of fault localization in time domain protection. In particular, a regression model for estimating fault location was trained with Laplace regularization” Here, Chakravorty establishes the model explicitly as a regression model.
None of Chakravorty, Watanabe, or Szmulewicz appear to explicitly teach “a determination unit configured to determine the priorities;”,
However in an analogous system, Sawada teaches “a determination unit configured to determine the priorities;”
See Sawada in paragraph [0061] describing, “The score calculation unit learning block 321 includes a determination unit 3211, a score calculation unit 3212, and a learning unit 3213. The determination unit 3211 is one example of the aforementioned determination unit 11. The score calculation unit 3212 is a model that calculates a score for the area as a priority for selecting the detection candidate area, that is, a processing module. In other words, the score calculation unit 3212 calculates the score that indicates the degree of the detection target for the candidate area using set parameters.” Here, Sawada shows a determination unit in combination with the score calculation unit in a learning block, the calculation unit determines priorities here by calculating a score for an area as a priority for selecting the detection candidate area which is being interpreted herein as determining priorities.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Chakravorty, Watanabe, and Szmulewicz with the teachings of Sawada by using Chakravorty’s teachings of regularization techniques in regression model, Watanabe’s teachings of an energy operation apparatus that includes a demand predictor using regression models, and Szmulewicz’s teachings of an objective function with a regularization term and incorporate with Sawada’s teachings of an apparatus that includes a determination unit and a learning unit.
One of ordinary skill in the art would be motivated to do so because by integrating Sawada’s frameworks into the methods of Chakravorty, Watanabe, and Szmulewicz, one of ordinary skill in the art would bring a “an image processing apparatus, a system, a method, and a program for improving a degree of accuracy of recognizing an image of one detection target from a set of images captured by a plurality of different modals.” (Sawada, paragraph [0026]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HASSAN R SESAY whose telephone number is (571)272-8493. The examiner can normally be reached Monday-Friday 8am-5pm.
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, Usmaan Saeed can be reached at (571) 272-4046. 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.
/HASSAN RAMADAN SESAY/Examiner, Art Unit 2146 /USMAAN SAEED/Supervisory Patent Examiner, Art Unit 2146
Prosecution Insights