Prosecution Insights
Last updated: November 20, 2026
Application No. 18/684,043

LINEAR MODEL VALIDATION

Non-Final OA §101§103
Filed
Feb 15, 2024
Priority
Aug 17, 2021 — EU 21191785.1 +1 more
Examiner
STANLEY, JEREMY L
Art Unit
2127
Tech Center
2100 — Computer Architecture & Software
Assignee
BASF SE
OA Round
1 (Non-Final)
50%
Grant Probability
Moderate
1-2
OA Rounds
6m
Est. Remaining
88%
With Interview

Examiner Intelligence

Grants 50% of resolved cases
50%
Career Allowance Rate
150 granted / 299 resolved
-4.8% vs TC avg
Strong +38% interview lift
Without
With
+38.1%
Interview Lift
resolved cases with interview
Typical timeline
3y 3m
Avg Prosecution
20 currently pending
Career history
329
Total Applications
across all art units

Statute-Specific Performance

§101
11.0%
-29.0% vs TC avg
§103
53.9%
+13.9% vs TC avg
§102
13.9%
-26.1% vs TC avg
§112
16.7%
-23.3% vs TC avg
Black line = Tech Center average estimate • Based on career data from 299 resolved cases

Office Action

§101 §103
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 . This action is responsive to the Application filed on February 15, 2024. Claims 1-15 are pending in the case. Claims 1 and 12 are the independent claims. This action is non-final. 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. Claim 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 the claims are directed to software per se; i.e., the claim recites a computer program product comprising program code means for causing a processor to carry out a computer-implemented method, which can be implemented entirely in software and therefore is not a machine, an article of manufacture, a process, or a composition of matter as contemplated ty 35 U.S.C. 101. Examiner respectfully suggests amending the claims to clarify that the claimed invention is not purely software. Claim 15 is rejected under 35 USC 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 the claims, under the broadest reasonable interpretation, are directed to transitory computer-readable media. Such media is not considered tangibly embodied and therefore does not therefore is not a machine, an article of manufacture, a process, or a composition of matter as contemplated ty 35 U.S.C. 101. Examiner respectfully suggests amending claim 20 to be drawn to a “non-transitory computer-readable medium. Claims 1-10, 12, 14, and 15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea (mental steps) without significantly more. This judicial exception is not integrated into a practical application because any additional elements amount to implementing the abstract idea on a generic computer. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Regarding independent claims 1 and 12, and relying on the evaluation flowchart in MPEP 2106: Step 1 (Is the claim to a process, machine, manufacture, or composition of matter?): Yes. Claim 1 is a system (machine). Claim 12 is a method (process). Step 2a Prong One (Does the claim recite an abstract idea?): Yes. Claims 1 and 12 recite: determining predicted data based on processing the input data by the linear model, wherein the predicted data includes predicted values for the different explanatory values (a mental process of determination, such as a human mentally determining predicted data including explanatory values); determining residual data based on a difference between the predicted data and the input data, wherein the residual data includes residual values for the different explanatory values determined based on the difference between the response values and their corresponding predicted values (a mental process involving performing a mathematical calculation, with or without the aid of pen and paper, such as a human mentally performing determining residual data based on the difference between response values and corresponding predicted values); determining whether the set of validation data fulfills a validation condition, wherein the validation condition is a homoscedasticity condition if the set of validation data includes homoscedasticity validation data or a normality condition if the set of validation data includes normality validation data (a mental process of evaluation, such as a human mentally evaluating whether the set of validation data fulfills a validation condition); determining…whether the set of validation data fulfills the validation condition (a mental process of evaluation, such as a human mentally determining whether the validation data fulfills the validation condition); determining whether the linear model is valid based on the output (a mental process of evaluation such as a human mentally determining that a model is valid based on output/results of the determination regarding the validation condition). Under the broadest reasonable interpretation, these steps may be performed mentally, using mental observation and mental determination, including by a human using a physical aid such as pen and paper, including a human mentally performing observations and mentally performing mathematical calculations, and therefore correspond to the Mental Processes grouping. Step 2a Prong Two (Does the claim recite additional elements that integrate the judicial exception into a practical application?): No. Claims 1 and 12 additionally recite: the system comprising a communication interface and a processor, wherein the communication interface is configured for…wherein the processor is configured for validating the linear model by performing the steps… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). receiving input data which includes response values for different explanatory values (insignificant extra-solution activity of transmitting data over a network as discussed in MPEP 2106.05(g)); generating a set of validation data based on the residual data, wherein the set of validation data includes homoscedasticity validation data or normality validation data (insignificant extra-solution activity as discussed in MPEP 2106.05(g), with respect to the generating limitation, and a field of use and technological environment as discussed in MPEP 2106.05(h) with respect to the validation data including homoscedasticity or normality validation data); providing a binary classifier for determining… wherein the binary classifier is a data driven model trained based on training sets of training validation data fulfilling the validation condition and training sets of training validation data not fulfilling the validation condition, such that the binary classifier outputs that the validation condition is fulfilled or not fulfilled depending on the set of validation data which is provided as input to the binary classifier, determining by the binary classifier…based on the output of the binary classifier (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Therefore, in view of the considerations set forth in MPEP 2106.04(d), 2106.05(a)-(c) and (e)-(h), the additional elements as disclosed above alone or in combination do not integrate the judicial exception into a practical application as they are mere insignificant extra solution activity, combined with implementing the abstract idea using generic computer components. Step 2b (Does the claim recite additional elements that amount to siqnificantly more than the judicial exception): No. Relying on the same analysis as Step 2a Prong Two (see MPEP 2106.05.I.A: Limitations that the courts have found not to be enough to qualify as “significantly more” when recited in a claim with a judicial exception include:…Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, e.g., a limitation indicating that a particular function such as creating and maintaining electronic records is performed by a computer, as discussed in Alice Corp., 573 U.S. at 225-26, 110 USPQ2d at 1984 (see MPEP 2106.05(f));…Simply appending well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception...; Adding insignificant extra-solution activity to the judicial exception, as discussed in MPEP 2106.05(g);…)), claims 1 and 12 do not recite any additional elements that amount to significantly more than the abstract idea. As discussed above, Claims 1 and 12 additionally recite: the system comprising a communication interface and a processor, wherein the communication interface is configured for…wherein the processor is configured for validating the linear model by performing the steps… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). receiving input data which includes response values for different explanatory values (insignificant extra-solution activity of transmitting data over a network as discussed in MPEP 2106.05(g), such as necessary data gathering, which can be reevaluated to include well-understood, routine, conventional activity such as receiving data over a network, retrieving information in memory, etc. discussed in MPEP 2106.05(d)); generating a set of validation data based on the residual data, wherein the set of validation data includes homoscedasticity validation data or normality validation data (insignificant extra-solution activity as discussed in MPEP 2106.05(g), such as necessary data outputting, with respect to the generating limitation, which can be reevaluated to include well-understood, routine, conventional activity such as transmitting data over a network, electronic recordkeeping, storing information in memory, etc. discussed in MPEP 2106.05(d), and a field of use and technological environment as discussed in MPEP 2106.05(h) with respect to the validation data including homoscedasticity or normality validation data); providing a binary classifier for determining… wherein the binary classifier is a data driven model trained based on training sets of training validation data fulfilling the validation condition and training sets of training validation data not fulfilling the validation condition, such that the binary classifier outputs that the validation condition is fulfilled or not fulfilled depending on the set of validation data which is provided as input to the binary classifier, determining by the binary classifier…based on the output of the binary classifier (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). The additional elements as discussed above, in combination with the abstract idea, are not sufficient to amount to significantly more than the judicial exception as they are well, understood, routine and conventional activity as disclosed in combination with generic computer functions and components used to implement the abstract idea. Regarding dependent claim 2: Step 2a Prong One: incorporates the rejection of claim 1. The claim additionally recites the linear model is determined to be valid (a mental process of determination). Step 2a Prong Two: the claims additionally recite wherein the processor is further configured for providing the linear model, if the linear model is determined to be valid (insignificant extra-solution activity as discussed in MPEP 2106.05(g) and mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Step 2b: the claims additionally recite wherein the processor is further configured for providing the linear model, if the linear model is determined to be valid (insignificant extra-solution activity as discussed in MPEP 2106.05(g), such as necessary data outputting, which can be reevaluated to include well-understood, routine, conventional activity such as transmitting data over a network, electronic recordkeeping, storing information in memory, etc. discussed in MPEP 2106.05(d), and mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Regarding dependent claim 3: Step 2a Prong One: incorporates the rejection of claim 1; the claim further recites adapting the linear model to an adapted linear model and…performing the steps performed for validating the linear model on the adapted linear model (a mental process involving performing a mathematical calculation, with or without the aid of pen and paper, such as a human mentally determining how to adapt the linear model and performing the mental validation steps as discussed above with respect to the independent claims). Step 2a Prong Two: the claim further recites wherein the processor is further configured for… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Step 2b: the claim further recites wherein the processor is further configured for… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Regarding dependent claim 4: Step 2a Prong One: incorporates the rejection of claim 3; the claim further recites adapting the linear model based on transforming the input data (a mental process involving performing a mathematical calculation, with or without the aid of pen and paper, such as a human mentally determining how to adapt the linear model based on transforming input data). Step 2a Prong Two: the claim further recites wherein the processor is further configured for… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)) Step 2b: the claim further recites wherein the processor is further configured for… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)) Regarding dependent claim 5: Step 2a Prong One: incorporates the rejection of claim 3; the claim further recites iteratively adapting the linear model until the adapted linear model is determined to be valid (a mental process of evaluation, such as a human mentally determining iterative adaptations of the linear model until it is determined to be valid). Step 2a Prong Two: the claim further recites wherein the processor is further configured for… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)) Step 2b: the claim further recites wherein the processor is further configured for… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)) Regarding dependent claim 6: Step 2a Prong One: incorporates the rejection of claim 1. Step 2a Prong Two: the claims additionally recite wherein the response values are measured plant yield values and the different explanatory values relate to different fields on which the plants grow (a field of use and technological environment as discussed in MPEP 2106.05(h)). Step 2b: the claims additionally recite wherein the response values are measured plant yield values and the different explanatory values relate to different fields on which the plants grow (a field of use and technological environment as discussed in MPEP 2106.05(h)). Regarding dependent claim 7: Step 2a Prong One: incorporates the rejection of claim 6. The claim additionally recites the linear model is determined to be valid (a mental process of determination). Step 2a Prong Two: the claims additionally recite wherein the system is configured for controlling a growth of the plants in the different fields based on the linear model (a field of use and technological environment as discussed in MPEP 2106.05(h)). Step 2b: the claims additionally recite wherein the system is configured for controlling a growth of the plants in the different fields based on the linear model (a field of use and technological environment as discussed in MPEP 2106.05(h)). Regarding dependent claim 8: Step 2a Prong One: incorporates the rejection of claim 1. The claims further recite determining whether the respective set of validation data fulfills a respective validation condition…determining whether the linear model is valid based on the output of the at least two binary classifiers (a mental process of determining whether validation data fulfills respective validation conditions and whether the linear model is valid based on an output). Step 2a Prong Two: the claim additionally recites: wherein the processor is configured for… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)) generating at least two different sets of validation data based on the residual data and for providing at least two binary classifiers (insignificant extra-solution activity as discussed in MPEP 2106.05(g)) binary classifiers configured for determining…wherein the processor is configured for determining…( mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Step 2b: the claim additionally recites: wherein the processor is configured for… (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)) generating at least two different sets of validation data based on the residual data and for providing at least two binary classifiers (insignificant extra-solution activity as discussed in MPEP 2106.05(g), such as necessary data outputting, which can be reevaluated to include well-understood, routine, conventional activity such as transmitting data over a network, electronic recordkeeping, storing information in memory, etc. discussed in MPEP 2106.05(d)) binary classifiers configured for determining…wherein the processor is configured for determining…( mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Regarding dependent claim 9: Step 2a Prong One: incorporates the rejection of claim 1. Step 2a Prong Two: the claim additionally recites wherein the binary classifier is a trained convolutional neural network (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Step 2b: the claim additionally recites wherein the binary classifier is a trained convolutional neural network (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Regarding dependent claim 10: Step 2a Prong One: incorporates the rejection of claim 9. Step 2a Prong Two: the claim additionally recites wherein the convolutional neural network has at least 4 convolutional layers (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Step 2b: the claim additionally recites wherein the convolutional neural network has at least 4 convolutional layers (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Regarding dependent claim 14: Step 2a Prong One: incorporates the rejection of claim 12. Step 2a Prong Two: the claim additionally recites A computer program product for validating a linear model, wherein the computer program product comprises program code means for causing a processor to carry out the computer-implemented method according to claim 12, when the computer program product is run on the processor (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Step 2b: the claim additionally recites A computer program product for validating a linear model, wherein the computer program product comprises program code means for causing a processor to carry out the computer-implemented method according to claim 12, when the computer program product is run on the processor (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Regarding dependent claim 15: Step 2a Prong One: incorporates the rejection of claim 14. Step 2a Prong Two: the claim additionally recites The computer readable medium having stored the computer program product of claim 14 (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Step 2b: the claim additionally recites The computer readable medium having stored the computer program product of claim 14 (mere instructions to apply the exception using generic computer components as discussed in MPEP 2106.05(f)). Therefore, in view of the considerations set forth in MPEP 2106.04(d), 2106.05(a)-(c) and (e)-(h), the additional elements as recited in the dependent claims discussed above alone or in combination do not integrate the judicial exception into a practical application as they are mere insignificant extra solution activity, combined with implementing the abstract idea using generic computer components, and limitations describing a field of use or technological environment. The additional elements as discussed above, in combination with the abstract idea, are not sufficient to amount to significantly more than the judicial exception as they are well, understood, routine and conventional activity as disclosed in combination with generic computer functions and components used to implement the abstract idea, and limitations describing a field of use or technological environment. Claim Rejections – 35 USC § 103 The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows: 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. This application currently names joint inventors. In considering patentability of the claims under pre-AIA 35 U.S.C. 103(a), the examiner presumes that the subject matter of the various claims was commonly owned at the time any inventions covered therein were made absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and invention dates of each claim that was not commonly owned at the time a later invention was made in order for the examiner to consider the applicability of pre-AIA 35 U.S.C. 103(c) and potential pre-AIA 35 U.S.C. 102€, (f) or (g) prior art under pre-AIA 35 U.S.C. 103(a). Claims 1, 8, 12, 14, and 15 are rejected under 35 U.S.C. 103 as being unpatentable over A. Gosiewska and P. Biecek. “auditor: an R Package for Model-Agnostic Visual Validation and Diagnostics.” ArXiv.org. May 26, 2020. Retrieved from the Internet: https://arxiv.org/pdf/1809.07763v4. (hereinafter “Gosiewsaka”) in view of M. Simic. “Testing for normality with neural networks.” Neural Computing and Applications. July 2021. Retreived from the Internet: https://arxiv.org/pdf/1809.07763v4. (hereinafter “Simic”). With respect to claim 1, Gosiewska teaches a system for validating a linear model, the system comprising a communication interface and a processor, wherein the communication interface is configured for receiving input data which includes response values for different explanatory values, and wherein the processor is configured for validating the linear model (e.g. abstract, machine learning models, software; presented methods implemented in the auditor package for R (i.e. statistical software); page 2, fifth paragraph, software dedicated to model auditing (i.e. where such software would be implemented using a processor and relevant interfaces); page 2 and pages 4-6, section 2.2, describing approaches for diagnostics/validation of linear models; page 8, section 3, the auditor package works for any predictive model which returns a numeric value, and offers a consistent grammar of model validation; page 9, Fig. 9, showing receipt of the model data) by performing the steps: determining predicted data based on processing the input data by the linear model, wherein the predicted data includes predicted values for the different explanatory values (e.g. page 1, section 1, introduction, predictive modeling uses mathematical and computational methods to forecast outcomes; page 2, footnote, value predicted by model), determining residual data based on a difference between the predicted data and the input data, wherein the residual data includes residual values for the different explanatory values determined based on the difference between the response values and their corresponding predicted values (e.g. page 2, footnote, indicating that the residual of an observation is the difference between observed value and value predicted by a model; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots), generating a set of validation data based on the residual data, wherein the set of validation data includes homoscedasticity validation data or normality validation data (e.g. page 1, section 1, indicating that model diagnostic objective 3 includes examination of problems relating to the model by analyzing distributions of residuals, in particular problems with bias, heteroscedasticity, and autocorrelation; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots; page 6, first paragraph, checking assumptions of linear regression model including normality of residual distribution and homoscedasticity of the variance; tests for normality and heteroscedasticity; page 15, final paragraph, continuing on page 16, describing a tool for assessing homoscedasticity of variance which assigns a score indicating whether residuals are homoscedastic or heteroscedastic), providing a tool for determining whether the set of validation data fulfills a validation condition, wherein the validation condition is a homoscedasticity condition if the set of validation data includes homoscedasticity validation data or a normality condition if the set of validation data includes normality validation data (e.g. page 1, section 1, indicating that model diagnostic objective 3 includes examination of problems relating to the model by analyzing distributions of residuals, in particular problems with bias, heteroscedasticity, and autocorrelation; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots; page 6, first paragraph, checking assumptions of linear regression model including normality of residual distribution and homoscedasticity of the variance; tests for normality and heteroscedasticity; page 15, final paragraph, continuing on page 16, describing a tool for assessing homoscedasticity of variance which assigns a score indicating whether residuals are homoscedastic or heteroscedastic; page 16, section 4.1.6, utilizing residual density plot to determine whether linear model residuals are normally distributed, which is one of the assumptions of linear regression; i.e. various software tools are provided for checking linear model assumptions including normality and homoscedasticity, in order to validate the model), determining by a tool whether the set of validation data fulfills the validation condition (e.g. page 1, section 1, indicating that model diagnostic objective 3 includes examination of problems relating to the model by analyzing distributions of residuals, in particular problems with bias, heteroscedasticity, and autocorrelation; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots; page 6, first paragraph, checking assumptions of linear regression model including normality of residual distribution and homoscedasticity of the variance; tests for normality and heteroscedasticity; page 15, final paragraph, continuing on page 16, describing a tool for assessing homoscedasticity of variance which assigns a score indicating whether residuals are homoscedastic or heteroscedastic; page 16, section 4.1.6, utilizing residual density plot to determine whether linear model residuals are normally distributed, which is one of the assumptions of linear regression; i.e. the provided software tools are utilized to determine whether the linear model assumptions are met or not, where these are utilized as validation conditions for the model), and determining whether the linear model is valid based on the output of the tool (e.g. page 1, section 1, indicating that model diagnostic objective 3 includes examination of problems relating to the model by analyzing distributions of residuals, in particular problems with bias, heteroscedasticity, and autocorrelation; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots; page 6, first paragraph, checking assumptions of linear regression model including normality of residual distribution and homoscedasticity of the variance; tests for normality and heteroscedasticity; page 15, final paragraph, continuing on page 16, describing a tool for assessing homoscedasticity of variance which assigns a score indicating whether residuals are homoscedastic or heteroscedastic; page 16, section 4.1.6, utilizing residual density plot to determine whether linear model residuals are normally distributed, which is one of the assumptions of linear regression; i.e. the provided software tools determine whether the model assumptions are valid or not, in order to determine whether the model is valid). Gosiewska does not explicitly disclose providing a binary classifier for determining whether the set of validation data fulfills a validation condition, wherein the validation condition is a homoscedasticity condition if the set of validation data includes homoscedasticity validation data or a normality condition if the set of validation data includes normality validation data, and wherein the binary classifier is a data driven model trained based on training sets of training validation data fulfilling the validation condition and training sets of training validation data not fulfilling the validation condition, such that the binary classifier determines whether the set of validation data fulfills the validation condition and outputs that the validation condition is fulfilled or not fulfilled depending on the set of validation data which is provided as input to the binary classifier. However, Simic teaches providing a binary classifier for determining whether the set of validation data fulfills a validation condition, wherein the validation condition is a homoscedasticity condition if the set of validation data includes homoscedasticity validation data or a normality condition if the set of validation data includes normality validation data, and wherein the binary classifier is a data driven model trained based on training sets of training validation data fulfilling the validation condition and training sets of training validation data not fulfilling the validation condition, such that the binary classifier outputs that the validation condition is fulfilled or not fulfilled depending on the set of validation data which is provided as input to the binary classifier (e.g. page 16282, second column, section 4, treating problem of testing for normality as a binary-classification problem; assigning sample label 1 if following a normal distribution and label 0 if following a non-normal distribution; neural network trained on dataset consisting of normal and non-normal samples of different sizes as with any other binary classification problem; output of neural network is a vector, the value of which is interpreted as the probability that it belongs to class 1, i.e. comes from a normal distribution). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska and Simic in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution), to incorporate the teachings of Simic (directed to testing for normality with neural networks) to include the capability to provide, a neural network performing binary classification for normality for the purpose of determining whether the validation data meets the validation condition, such as a normality condition. One of ordinary skill would have been motivated to perform such a modification in order to provide a neural network with very high potential for use in everyday practice of statistics, data analysis, and machine learning as described in Simic (abstract). With respect to claim 12, Gosiewska teaches a computer implemented method for validating a linear model, comprising: receiving input data which includes response values for different explanatory values, and validating the linear model by performing the steps (e.g. page 2 and pages 4-6, section 2.2, describing approaches for diagnostics/validation of linear models; page 8, section 3, the auditor package works for any predictive model which returns a numeric value, and offers a consistent grammar of model validation; page 9, Fig. 9, showing receipt of the model data): determining predicted data based on processing the input data by the linear model, wherein the predicted data includes predicted values for the different explanatory values (e.g. page 1, section 1, introduction, predictive modeling uses mathematical and computational methods to forecast outcomes; page 2, footnote, value predicted by model), determining residual data based on a difference between the predicted data and the input data, wherein the residual data includes residual values for the different explanatory values determined based on the difference between the response values and their corresponding predicted values (e.g. page 2, footnote, indicating that the residual of an observation is the difference between observed value and value predicted by a model; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots), generating a set of validation data based on the residual data, wherein the set of validation data includes homoscedasticity validation data or normality validation data (e.g. page 1, section 1, indicating that model diagnostic objective 3 includes examination of problems relating to the model by analyzing distributions of residuals, in particular problems with bias, heteroscedasticity, and autocorrelation; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots; page 6, first paragraph, checking assumptions of linear regression model including normality of residual distribution and homoscedasticity of the variance; tests for normality and heteroscedasticity; page 15, final paragraph, continuing on page 16, describing a tool for assessing homoscedasticity of variance which assigns a score indicating whether residuals are homoscedastic or heteroscedastic), providing a tool for determining whether the set of validation data fulfills a validation condition, wherein the validation condition is a homoscedasticity condition if the set of validation data includes homoscedasticity validation data or a normality condition if the set of validation data includes normality validation data (e.g. page 1, section 1, indicating that model diagnostic objective 3 includes examination of problems relating to the model by analyzing distributions of residuals, in particular problems with bias, heteroscedasticity, and autocorrelation; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots; page 6, first paragraph, checking assumptions of linear regression model including normality of residual distribution and homoscedasticity of the variance; tests for normality and heteroscedasticity; page 15, final paragraph, continuing on page 16, describing a tool for assessing homoscedasticity of variance which assigns a score indicating whether residuals are homoscedastic or heteroscedastic; page 16, section 4.1.6, utilizing residual density plot to determine whether linear model residuals are normally distributed, which is one of the assumptions of linear regression; i.e. various software tools are provided for checking linear model assumptions including normality and homoscedasticity, in order to validate the model), determining by the binary classifier whether the set of validation data fulfils the validation condition (e.g. page 1, section 1, indicating that model diagnostic objective 3 includes examination of problems relating to the model by analyzing distributions of residuals, in particular problems with bias, heteroscedasticity, and autocorrelation; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots; page 6, first paragraph, checking assumptions of linear regression model including normality of residual distribution and homoscedasticity of the variance; tests for normality and heteroscedasticity; page 15, final paragraph, continuing on page 16, describing a tool for assessing homoscedasticity of variance which assigns a score indicating whether residuals are homoscedastic or heteroscedastic; page 16, section 4.1.6, utilizing residual density plot to determine whether linear model residuals are normally distributed, which is one of the assumptions of linear regression; i.e. the provided software tools are utilized to determine whether the linear model assumptions are met or not, where these are utilized as validation conditions for the model), and determining whether the linear model is valid based on the output of the binary classifier (e.g. page 1, section 1, indicating that model diagnostic objective 3 includes examination of problems relating to the model by analyzing distributions of residuals, in particular problems with bias, heteroscedasticity, and autocorrelation; page 2, fourth paragraph, indicating that diagnostics analysis for linear models is typically performed by extracting raw, studentized, deviance, or Pearson residuals and examining residual plots; page 6, first paragraph, checking assumptions of linear regression model including normality of residual distribution and homoscedasticity of the variance; tests for normality and heteroscedasticity; page 15, final paragraph, continuing on page 16, describing a tool for assessing homoscedasticity of variance which assigns a score indicating whether residuals are homoscedastic or heteroscedastic; page 16, section 4.1.6, utilizing residual density plot to determine whether linear model residuals are normally distributed, which is one of the assumptions of linear regression; i.e. the provided software tools determine whether the model assumptions are valid or not, in order to determine whether the model is valid). Gosiewska does not explicitly disclose providing a binary classifier for determining whether the set of validation data fulfills a validation condition, wherein the validation condition is a homoscedasticity condition if the set of validation data includes homoscedasticity validation data or a normality condition if the set of validation data includes normality validation data, and wherein the binary classifier is a data driven model trained based on training sets of training validation data fulfilling the validation condition and training sets of training validation data not fulfilling the validation condition, such that the binary classifier determines whether the set of validation data fulfills the validation condition and outputs that the validation condition is fulfilled or not fulfilled depending on the set of validation data which is provided as input to the binary classifier. However, Simic teaches providing a binary classifier for determining whether the set of validation data fulfills a validation condition, wherein the validation condition is a homoscedasticity condition if the set of validation data includes homoscedasticity validation data or a normality condition if the set of validation data includes normality validation data, and wherein the binary classifier is a data driven model trained based on training sets of training validation data fulfilling the validation condition and training sets of training validation data not fulfilling the validation condition, such that the binary classifier outputs that the validation condition is fulfilled or not fulfilled depending on the set of validation data which is provided as input to the binary classifier (e.g. page 16282, second column, section 4, treating problem of testing for normality as a binary-classification problem; assigning sample label 1 if following a normal distribution and label 0 if following a non-normal distribution; neural network trained on dataset consisting of normal and non-normal samples of different sizes as with any other binary classification problem; output of neural network is a vector, the value of which is interpreted as the probability that it belongs to class 1, i.e. comes from a normal distribution). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska and Simic in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution), to incorporate the teachings of Simic (directed to testing for normality with neural networks) to include the capability to provide, a neural network performing binary classification for normality for the purpose of determining whether the validation data meets the validation condition, such as a normality condition. One of ordinary skill would have been motivated to perform such a modification in order to provide a neural network with very high potential for use in everyday practice of statistics, data analysis, and machine learning as described in Simic (abstract). With respect to claim 14, Gosiewska in view of Simic teaches a computer program product for validating a linear model, wherein the computer program product comprises program code means for causing a processor to carry out the computer-implemented method according to claim 12, when the computer program product is run on the processor (e.g. Gosiewska abstract, machine learning models, software; presented methods implemented in the auditor package for R (i.e. statistical software); page 2, fifth paragraph, software dedicated to model auditing). With respect to claim 15, Gosiewska in view of Simic teaches the computer readable medium having stored the computer program product of claim 14 (e.g. Gosiewska abstract, machine learning models, software; presented methods implemented in the auditor package for R (i.e. statistical software); page 2, fifth paragraph, software dedicated to model auditing). With respect to claim 8, Gosiewska in view of Simic teaches all of the limitations of claim 1 as previously discussed, and further teaches wherein the processor is configured for generating at least two different sets of validation data based on the residual data and for providing at least two tools each configured for determining whether the respective set of validation data fulfills a respective validation condition and wherein the processor is configured for determining whether the linear model is valid based on the output of the at least two tools (e.g. page 4, first paragraph, describing a variety of tests for normality of residual distribution and homoscedasticity of the variance; pages 14-16, Section 4.1.5, describing scale location plot to visualize variance of the residuals, where the presence of a trend suggest variance depends on fitted values and is against the assumption of homoscedasticity; Peak test is tool for assessing homoscedasticity of variance; Peak score correlated with residuals; page 16, section 4.1.6., describing residual density plot as detecting incorrect behavior of residuals; simple linear model residuals should be normally distributed; normal distribution is one of the assumptions of simple linear regression; i.e. the validation data is provided in at least two forms, such as a scale location plot for visualizing variance of residuals and a residual density plot for detecting incorrect behavior of the residuals, and associated tools are provided for processing each to determine model validity according to corresponding conditions, such as validation of the assumption of normality of residual distribution and homoscedasticity of the variance). Gosiewska does not explicitly disclose that the tools are at least two binary classifiers. However, Simic teaches that the tools are at least two binary classifiers (e.g. page 16284, second column, setting DBNN parameters such as decision threshold in advance, such that DBNN can be used just as any standard statistical test; pages 16286-16287,section 6.4, indicating that the DBNN can be set with multiple decision threshold values, such that different variants of the DBNN with different thresholds act as different classifiers; i.e. the binary classifier/DBNN can be configured with different threshold corresponding to different statistical tests, such that the different configurations act as multiple different binary classifiers). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska and Simic in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution), to incorporate the teachings of Simic (directed to testing for normality with neural networks) to include the capability to provide multiple configurations of a neural network performing binary classification for a variety of different statistical tests for the purpose of determining whether the validation data meets the validation condition, such as a normality condition. One of ordinary skill would have been motivated to perform such a modification in order to provide a neural network with very high potential for use in everyday practice of statistics, data analysis, and machine learning as described in Simic (abstract). Claims 2-5 are rejected under 35 U.S.C. 103 as being unpatentable over Gosiewska in view of M. Simic, further in view of Vengalayapalli Kichagari (US 20200210856 A1). With respect to claim 2, Gosiewska in view of Simic teaches all of the limitations of claim 1 as previously discussed. Gosiewska and Simic do not explicitly disclose wherein the processor is further configured for providing the linear model, if the linear model is determined to be valid. However, Vengalayapalli Kichagari teaches wherein the processor is further configured for providing the linear model, if the linear model is determined to be valid (e.g. paragraph 0142, indicating that predictive model may be a linear regression model, etc.; paragraph 0145, Fig. 6, step 614, performing predictive model validation including residual diagnosis allowing for checking of normality assumption, testing of assumption of homoscedasticity, etc. if residual diagnosis valid, variables enter into the back testing; paragraph 0146, if residuals valid, multi collinearity checked between variables and evaluated as to whether it is tolerable; step 630, checking if predictive model is valid; paragraph 0147, if predictive model is valid, in step 632, prediction is performed on live data; paragraph 0149, if result of back testing is fitting the requirement, predictive model is utilized to predicted from live data entered into the system; i.e. once the model is determined to be valid, it is provided for performing predictions on live data). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, and Vengalayapalli Kichagari in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution) and Simic (directed to testing for normality with neural networks), to incorporate the teachings of Vengalayapalli Kichagari (directed to predictive model building and validation) to include the capability to provide the linear model, such as for performing predictions on live data, if it is determined to be valid. One of ordinary skill would have been motivated to perform such a modification in order to provide a model that is simple and has good predictability, and to allow analysis and validation of predictability of the model as described in Vengalayapalli Kichagari (paragraph 0120). With respect to claim 3, Gosiewska in view of Simic teaches all of the limitations of claim 1 as previously discussed. Gosiewska and Simic do not explicitly disclose wherein the processor is further configured for adapting the linear model to an adapted linear model and for performing the steps performed for validating the linear model on the adapted linear model. However, Vengalayapalli Kichagari teaches wherein the processor is further configured for adapting the linear model to an adapted linear model and for performing the steps performed for validating the linear model on the adapted linear model (e.g. paragraph 0145-0146, Fig. 6, if residual diagnosis is not valid, then the variables are transformed, then the variables are transformed, or outliers are removed, and the variables enter into the step of stepwise variable selection; if residuals found not valid, enabling transformation so as to remove outliers and thereafter going back to step 614; i.e. as shown in Fig. 6, where residuals are not valid following steps 618 and 620, transformation/outlier removal is performed, stepwise variable selection is performed again at step 614, and the model is then rebuilt/adapted and re-validated at steps 616-630). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, and Vengalayapalli Kichagari in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution) and Simic (directed to testing for normality with neural networks), to incorporate the teachings of Vengalayapalli Kichagari (directed to predictive model building and validation) to include the capability to adapt the linear model, such as by performing associated transformation and variable reselection, and subsequently rebuilding the model), and to perform validation of the adapted linear model. One of ordinary skill would have been motivated to perform such a modification in order to provide a model that is simple and has good predictability, and to allow analysis and validation of predictability of the model as described in Vengalayapalli Kichagari (paragraph 0120). With respect to claim 4, Gosiewska in view of Simic, further in view of Vengalayapalli Kichagari teaches all of the limitations of claim 3 as previously discussed, and Vengalayapalli Kichagari further teaches wherein the processor is configured for adapting the linear model based on transforming the input data (e.g. paragraph 0145-0146, Fig. 6, if residual diagnosis is not valid, then the variables are transformed, then the variables are transformed, or outliers are removed, and the variables enter into the step of stepwise variable selection; if residuals found not valid, enabling transformation so as to remove outliers and thereafter going back to step 614; i.e. as shown in Fig. 6, variables are the result of conversion of merged input data (step 608, analogous to transformation of input data) and the variables themselves are selected on a step-wise basis as input data for building the model (step 614), such that performing transformation of the variables and removal of outliers at following step 620, and then utilizing those transformed variables to perform step 614 and step 616 again is analogous to adapting the linear model based on transforming the input data). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, and Vengalayapalli Kichagari in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution) and Simic (directed to testing for normality with neural networks), to incorporate the teachings of Vengalayapalli Kichagari (directed to predictive model building and validation) to include the capability to adapt the linear model based on transforming the input data. One of ordinary skill would have been motivated to perform such a modification in order to provide a model that is simple and has good predictability, and to allow analysis and validation of predictability of the model as described in Vengalayapalli Kichagari (paragraph 0120). With respect to claim 5, Gosiewska in view of Simic, further in view of Vengalayapalli Kichagari teaches all of the limitations of claim 3 as previously discussed, and Vengalayapalli Kichagari further teaches wherein the processor is further configured for iteratively adapting the linear model until the adapted linear model is determined to be valid (e.g. paragraph 0145-0146, Fig. 6, if residual diagnosis is not valid, then the variables are transformed, then the variables are transformed, or outliers are removed, and the variables enter into the step of stepwise variable selection; if residuals found not valid, enabling transformation so as to remove outliers and thereafter going back to step 614; i.e. as shown in Fig. 6, model is checked for validity (such as at steps 620 and 630), and if the model is not found to be valid, the process of building the model and checking validity again is continually repeated until the model is finally found to be valid, analogous to iteratively adapting the linear model until the adapted linear model is determined to be valid). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, and Vengalayapalli Kichagari in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution) and Simic (directed to testing for normality with neural networks), to incorporate the teachings of Vengalayapalli Kichagari (directed to predictive model building and validation) to include the capability to iteratively adapt the linear model until it is determined to be valid. One of ordinary skill would have been motivated to perform such a modification in order to provide a model that is simple and has good predictability, and to allow analysis and validation of predictability of the model as described in Vengalayapalli Kichagari (paragraph 0120). With respect to claim 13, Gosiewska in view of Simic teaches all of the limitations of claim 12 as previously discussed. Gosiewska in view of Simic does not explicitly disclose the computer implemented method including one or more of the steps: providing the linear model, if the linear model is determined to be valid, adapting the linear model to an adapted linear model, and performing the steps performed for validating the linear model on the adapted linear model, adapting the linear model based on transforming the input data, iteratively adapting the linear model until the adapted linear model is determined to be valid, measuring plant yield values as response values related to different fields on which the plants grow as explanatory values, controlling growth of the plants in the different fields based on the linear model, if the linear model is determined to be valid, generating at least two different sets of validation data based on the residual data, providing at least two binary classifiers each configured for determining whether the respective set of validation data fulfills a respective validation condition, determining whether the linear model is valid based on the output of the at least two binary classifiers, providing the binary classifier as a convolutional neural network, providing that the convolutional neural network has at least 4 convolutional layers, providing that the convolutional neural network has the following architectures: a first convolutional layer, a first max pooling layer, a second convolutional layer, a third convolutional layer, a second max pooling layer, a dropout layer, a fourth convolutional layer, a global max pooling layer, a first densely-connected layer, and a second densely-connected layer, and training the binary classifier based on training sets of training validation data fulfilling the validation condition and training sets of training validation data not fulfilling the validation condition, such that the binary classifier outputs that the validation condition is fulfilled or not fulfilled depending on the set of validation data which is provided as input to the binary classifier. However, Vengalayapalli Kichagari teaches the computer implemented method including one or more of the steps: providing the linear model, if the linear model is determined to be valid (e.g. paragraph 0142, indicating that predictive model may be a linear regression model, etc.; paragraph 0145, Fig. 6, step 614, performing predictive model validation including residual diagnosis allowing for checking of normality assumption, testing of assumption of homoscedasticity, etc. if residual diagnosis valid, variables enter into the back testing; paragraph 0146, if residuals valid, multi collinearity checked between variables and evaluated as to whether it is tolerable; step 630, checking if predictive model is valid; paragraph 0147, if predictive model is valid, in step 632, prediction is performed on live data; paragraph 0149, if result of back testing is fitting the requirement, predictive model is utilized to predicted from live data entered into the system; i.e. once the model is determined to be valid, it is provided for performing predictions on live data), adapting the linear model to an adapted linear model, and performing the steps performed for validating the linear model on the adapted linear model (e.g. paragraph 0145-0146, Fig. 6, if residual diagnosis is not valid, then the variables are transformed, then the variables are transformed, or outliers are removed, and the variables enter into the step of stepwise variable selection; if residuals found not valid, enabling transformation so as to remove outliers and thereafter going back to step 614; i.e. as shown in Fig. 6, where residuals are not valid following steps 618 and 620, transformation/outlier removal is performed, stepwise variable selection is performed again at step 614, and the model is then rebuilt/adapted and re-validated at steps 616-630), adapting the linear model based on transforming the input data (e.g. paragraph 0145-0146, Fig. 6, if residual diagnosis is not valid, then the variables are transformed, then the variables are transformed, or outliers are removed, and the variables enter into the step of stepwise variable selection; if residuals found not valid, enabling transformation so as to remove outliers and thereafter going back to step 614; i.e. as shown in Fig. 6, variables are the result of conversion of merged input data (step 608, analogous to transformation of input data) and the variables themselves are selected on a step-wise basis as input data for building the model (step 614), such that performing transformation of the variables and removal of outliers at following step 620, and then utilizing those transformed variables to perform step 614 and step 616 again is analogous to adapting the linear model based on transforming the input data), iteratively adapting the linear model until the adapted linear model is determined to be valid (e.g. paragraph 0145-0146, Fig. 6, if residual diagnosis is not valid, then the variables are transformed, then the variables are transformed, or outliers are removed, and the variables enter into the step of stepwise variable selection; if residuals found not valid, enabling transformation so as to remove outliers and thereafter going back to step 614; i.e. as shown in Fig. 6, model is checked for validity (such as at steps 620 and 630), and if the model is not found to be valid, the process of building the model and checking validity again is continually repeated until the model is finally found to be valid, analogous to iteratively adapting the linear model until the adapted linear model is determined to be valid), measuring plant yield values as response values related to different fields on which the plants grow as explanatory values, controlling growth of the plants in the different fields based on the linear model, if the linear model is determined to be valid, generating at least two different sets of validation data based on the residual data, providing at least two binary classifiers each configured for determining whether the respective set of validation data fulfills a respective validation condition, determining whether the linear model is valid based on the output of the at least two binary classifiers, providing the binary classifier as a convolutional neural network, providing that the convolutional neural network has at least 4 convolutional layers, providing that the convolutional neural network has the following architectures: a first convolutional layer, a first max pooling layer, a second convolutional layer, a third convolutional layer, a second max pooling layer, a dropout layer, a fourth convolutional layer, a global max pooling layer, a first densely-connected layer, and a second densely-connected layer, and training the binary classifier based on training sets of training validation data fulfilling the validation condition and training sets of training validation data not fulfilling the validation condition, such that the binary classifier outputs that the validation condition is fulfilled or not fulfilled depending on the set of validation data which is provided as input to the binary classifier. With respect to the limitations (method steps) above which Vengalayapalli Kichagari is not cited as teaching, Examiner notes that the claim only appears to require that the computer implemented method include one or more of the steps (limitations), and the citations with respect to the multiple recited steps taught by Vengalayapalli Kichagari are sufficient to meet the claim as currently written. Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, and Vengalayapalli Kichagari in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution) and Simic (directed to testing for normality with neural networks), to incorporate the teachings of Vengalayapalli Kichagari (directed to predictive model building and validation) to include the capability to iteratively adapt the linear model until it is determined to be valid. One of ordinary skill would have been motivated to perform such a modification in order to provide a model that is simple and has good predictability, and to allow analysis and validation of predictability of the model as described in Vengalayapalli Kichagari (paragraph 0120). Claims 6 and 7 are rejected under 35 U.S.C. 103 as being unpatentable over Gosiewska in view of M. Simic, further in view of A. Komarek, Ch. Thierfelder, P. Steward. “Conservation agriculture improves adaptive capacity of cropping systems to climate stress in Malawi.” Agricultural Systems, Volume 190, 2021, 103117, ISSN 0308-521X, https://doi.org/10.1016/j.agsy.2021.103117. (https://www.sciencedirect.com/science/article/pii/S0308521X21000706). (hereinafter referred to as “Komarek”). With respect to claim 6, Gosiewska in view of Simic teaches all of the limitations of claim 1 as previously discussed. Gosiewska and Simic do not explicitly disclose wherein the response values are measured plant yield values and the different explanatory values relate to different fields on which the plants grow. However, Komarek teaches wherein the response values are measured plant yield values and the different explanatory values relate to different fields on which the plants grow (e.g. page 2, second column, section 2.1.1, describing study as including, in each community, six trial replicates on six farmer’s fields, the fields divided into three adjacent plots, each plot containing one of three treatments; page 7, first column, section 2.4, linear mixed-effects models examining effect of each treatment on cropping-system yield; global model including all explanatory variables and their interactions suspected of affecting cropping-system yields, including a numerical variable for the total number of seasons a plot has been in the on-farm trials, variables for identification of associated community and farmer, etc.; i.e. the model input data includes cropping system yield data and multiple explanatory data values relating to the fields in which the crops grow (including number of years, community, and farmer associated with the different plots)). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, and Komarek in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution) and Simic (directed to testing for normality with neural networks), to incorporate the teachings of Komarek (directed to use of linear models to examine effects of explanatory variables on crop yields) to include the capability to provide, in the input data, response values which are measured plant yield values and different explanatory values which relate to different fields on which the plants grow. One of ordinary skill would have been motivated to perform such a modification in order to improve the adaptive capacity of cropping systems as described in Komarek (page 1, abstract) With respect to claim 7, Gosiewska in view of Simic teaches all of the limitations of claim 6 as previously discussed, and further teaches wherein the system is configured for controlling a growth of the plants in the different fields based on the linear model, if the linear model is determined to be valid (e.g. page 2, second column, first paragraph, statistical model using daily weather data as an explanatory variable as part of study; study provides alternative approach to studying weather and treatment interactions by using daily weather data over the growing season to calculate explanatory variables for climate conditions during the growing season; combining use of cropping system yields and climate stress indicators from daily weather data into one study; page 2, second column, section 2.1.1, describing study as including, in each community, six trial replicates on six farmer’s fields, the fields divided into three adjacent plots, each plot containing one of three treatments, and further describing the three treatments; page 7, first column, section 2.4, using linear model to examine the effect of each treatment on cropping system yield and how these treatment effects depend on precipitation balance and heat stress for maize; testing suspicion that variables in model equation affect cropping system yields in dataset; estimating models separately for each of three crop management strategies page 7, second column, first full paragraph, model selection/identification of suitable approximating model for predicting cropping-system yield; comparing models within each of three strategies, estimating retained models, generating estimated coefficients for retained models, and, for model assumptions, inspecting normality of residuals and homogeneity assumption; i.e. the model, once selected and validated as suitable, is developed and used as part of the study, including for controlling growth of plants/crops in different fields/plots according to three different treatments). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, and Komarek in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution) and Simic (directed to testing for normality with neural networks), to incorporate the teachings of Komarek (directed to use of linear models to examine effects of explanatory variables on crop yields) to include the capability to control growth of the plants in the different fields based on the linear model determined to be valid, i.e. by utilizing the model as part of a study of different methods for treating crops in different fields/plots including corresponding cropping system yields. One of ordinary skill would have been motivated to perform such a modification in order to improve the adaptive capacity of cropping systems as described in Komarek (page 1, abstract). Claims 9 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Gosiewska in view of M. Simic, further in view of K. Huang, M. Qiao, X. Liu, S. Liu, M. Dai. “Computer Vision and Metrics Learning for Hypothesis Testing: An Application of Q-Q Plot for Normality Test.” ArXiv.org. September 16, 2019. Retrieved from the Internet: https://arxiv.org/pdf/1901.07851. (hereinafter referred to as “Huang”). With respect to claim 9, Gosiewska in view of Simic teaches all of the limitations of claim 1 as previously discussed. Gosiewska and Simic do not explicitly disclose wherein the binary classifier is a trained convolutional neural network. However, Huang teaches wherein the binary classifier is a trained convolutional neural network (e.g. page 1, final paragraph, continuing on page 2, Q-Q plot commonly used in linear regression analysis to examine whether regression residuals are normally distributed; page 2, first and second full paragraphs, the graphical method can serve as effective tool in checking normality; using computer vision to automate this visualization task; investigating whether AI-based method can outperform existing normality tests developed by statisticians; page 3, first paragraph, quantifying similarity between Q-Q plots utilizing deep convolution neural network to extract Q-Q plot image representation features which map Q-Q plots to numerical vectors, conduct metric learning to quantify differences between two Q-Q plots by LMNN metric learning; similar to traditional normality test, we can computationally derive the distribution of this test statistic; page 4, section 2.2. discussing various convolutional neural networks for image classification; page 11, final paragraph, continuing on page 12, generalizing method to construct test statistics for other famous hypothesis testing in statistics based on other statistical visualizations, such as residual plots for linear regression analysis). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, and Huang in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution) and Simic (directed to testing for normality with neural networks), to incorporate the teachings of Huang (directed to computer vision and metrics learning for hypothesis testing applied to Q-Q plot for normality test, such as in examining normality in linear regression) to include the capability to implement the binary classifier (i.e. of Simic) as a trained convolutional neural network. One of ordinary skill would have been motivated to perform such a modification in order to provide machine learning test statistics that outperform traditional normality tests as described in Huang (page 1, abstract). With respect to claim 10, Gosiewska in view of Simic, further in view of Huang teaches all of the limitations of claim 9 as previously discussed, and Huang further teaches wherein the convolutional neural network has at least 4 convolutional layers (e.g. page 1, final paragraph, continuing on page 2, Q-Q plot commonly used in linear regression analysis to examine whether regression residuals are normally distributed; page 2, first and second full paragraphs, the graphical method can serve as effective tool in checking normality; using computer vision to automate this visualization task; investigating whether AI-based method can outperform existing normality tests developed by statisticians; page 3, first paragraph, quantifying similarity between Q-Q plots utilizing deep convolution neural network to extract Q-Q plot image representation features which map Q-Q plots to numerical vectors, conduct metric learning to quantify differences between two Q-Q plots by LMNN metric learning; similar to traditional normality test, we can computationally derive the distribution of this test statistic; page 4, section 2.2. discussing various convolutional neural networks for image classification, including VGG-Net, which consists of 16-19 layers; page 5, section 3.1.1, deep neural network structure can be LeNet, AlexNet, VGG, ResNet, or Inception; VGG-19 employed as main model; page 6, Fig. 2, showing the framework including the CNN architecture having at least 4 convolutional layers; page 11, final paragraph, continuing on page 12, generalizing method to construct test statistics for other famous hypothesis testing in statistics based on other statistical visualizations, such as residual plots for linear regression analysis). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, and Huang in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution) and Simic (directed to testing for normality with neural networks), to incorporate the teachings of Huang (directed to computer vision and metrics learning for hypothesis testing applied to Q-Q plot for normality test, such as in examining normality in linear regression) to include the capability to implement the binary classifier (i.e. of Simic) as a trained convolutional neural network having at least 4 convolutional layers. One of ordinary skill would have been motivated to perform such a modification in order to provide machine learning test statistics that outperform traditional normality tests as described in Huang (page 1, abstract). Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Gosiewska in view of M. Simic, further in view of Huang, further in view of J. Zhang, A. Liu, M. Gao, X. Chen, X. Zhang, X. Chen. "ECG-based multi-class arrhythmia detection using spatio-temporal attention-based convolutional recurrent neural network." Artificial Intelligence in Medicine, Volume 106, June 2020, 101856, ISSN 0933-3657, https://doi.org/10.1016/j.artmed.2020.101856. (https://www.sciencedirect.com/science/article/pii/S0933365719312606). (hereinafter referred to as "Zhang"). With respect to claim 11, Gosiewska in view of Simic, further in view of Huang teaches all of the limitations of claim 9 as previously discussed. Although Huang teaches use of a CNN architecture for residual plot analysis, such as based on VGGNet, Gosiewska, Simic, and Huang do not explicitly disclose wherein the convolutional neural network has the following architecture: a first convolutional layer, a first max pooling layer, a second convolutional layer, a third convolutional layer, a second max pooling layer, a dropout layer, a fourth convolutional layer, a global max pooling layer, a first densely-connected layer, and a second densely-connected layer. However, Zhang teaches wherein the convolutional neural network has the following architecture (e.g. page 3, section 3.2 and 3.2.1, describing the model architecture as shown in Figs. 1 and 2, the proposed model integrates CNN, spatio-temporal attention module, and GRU layer; CNN is inspired by VGGNet and applied to extract local features, and includes 5 convolutional blocks/13 1-D convolution layers, max-pooling layers, and dropout layers): a first convolutional layer (e.g. page 3, Fig. 2, CNN has at least a 1th convolutional block including convolutional layers), a first max pooling layer (e.g. page 3, Fig. 2, CNN has at least a 1th convolutional block including a max-pooling layer), a second convolutional layer (e.g. page 3, Fig. 2, CNN has at least a 2th convolutional block including convolutional layers), a third convolutional layer (e.g. page 3, Fig. 2, CNN has at least a 3th convolutional block including convolutional layers), a second max pooling layer (e.g. page 3, Fig. 2, CNN has at least a 2th or 3th convolutional block including max-pooling layers), a dropout layer (e.g. page 3, Fig. 2, CNN has at least a 1th, 2th, and 3th-5th convolutional blocks including dropout layers), a fourth convolutional layer (e.g. page 3, Fig. 2, CNN has at least a 3th-5th convolutional block including convolutional layers (i.e. a fourth convolutional block and corresponding layers)), a global max pooling layer (e.g. Fig. 1, CNN is combined with spatio-temporal attention modules; page 3, Fig. 3, showing that the spatial attention mechanism includes at least a global max pooling layer; moreover, as shown in Fig. 1, a global max pooling layer is also present in the final portion of the overall model), a first densely-connected layer (e.g. Fig. 1, CNN is combined with spatio-temporal attention modules; page 3, Fig. 3, showing that the spatial attention mechanism includes at least shared 2 dense layers), and a second densely-connected layer (e.g. Fig. 1, CNN is combined with spatio-temporal attention modules; page 3, Fig. 3, showing that the spatial attention mechanism includes at least shared 2 dense layers). Accordingly, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention having the teachings of Gosiewska, Simic, Huang, and Zhang in front of him to have modified the teachings of Gosiewska (directed to model-agnostic visual validation and diagnostics, including checking linear model assumptions such as normality of residual distribution), Simic (directed to testing for normality with neural networks), and Huang (directed to computer vision and metrics learning for hypothesis testing applied to Q-Q plot for normality test, such as in examining normality in linear regression), to incorporate the teachings of Zhang (directed to a spatio-temporal attention-based convolutional recurrent neural network) to include the capability to implement the binary classifier (i.e. of Simic) as a trained convolutional neural network having at least 4 convolutional layers (as taught by Huang, where the CNN may be implemented using VGGNet, etc.), and may further be implemented by including at least 2 max pooling layers, at least one dropout layer, at least one global max pooling layer, and at least 2 densely-connected layers. One of ordinary skill would have been motivated to perform such a modification in order to provide improved classification performance in models incorporating CNN architecture as described in Zhang (page 1, abstract). It is noted that any citation to specific pages, columns, lines, or figures in the prior art references and any interpretation of the references should not be considered to be limiting in any way. “The use of patents as references is not limited to what the patentees describe as their own inventions or to the problems with which they are concerned. They are part of the literature of the art, relevant for all they contain,” In re Heck, 699 F.2d 1331, 1332-33, 216 USPQ 1038, 1039 (Fed. Cir. 1983) (quoting in re Lemelson, 397 F.2d 1006, 1009, 158 USPQ 275, 277 (GCPA 1968)). Further, a reference may be relied upon for all that it would have reasonably suggested to one having ordinary skill the art, including nonpreferred embodiments. Merck & Co, v. Biocraft Laboratories, 874 F.2d 804, 10 USPQ2d 1843 (Fed. Cir.), cert, denied, 493 U.S. 975 (1989). See also Upsher-Smith Labs. v. Pamlab, LLC, 412 F,3d 1319, 1323, 75 USPQ2d 1213, 1215 (Fed. Cir, 2005): Celeritas Technologies Ltd. v. Rockwell International Corp., 150 F.3d 1354, 1361, 47 USPQ2d 1516, 1522-23 (Fed. Cir. 1998). Conclusion The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure. Any inquiry concerning this communication or earlier communications from the examiner should be directed to JEREMY L STANLEY whose telephone number is (469)295-9105. The examiner can normally be reached on Monday-Friday from 9:00 AM to 5:00 PM CST. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Abdullah Al Kawsar, can be reached at telephone number (571) 270-3169. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. Information regarding the status of an application may be obtained from Patent Center and the Private Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from Patent Center or Private PAIR. Status information for unpublished applications is available through Patent Center and Private PAIR for authorized users only. Should you have questions about access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). 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) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form. /JEREMY L STANLEY/ Primary Examiner, Art Unit 2127
Read full office action

Prosecution Timeline

Feb 15, 2024
Application Filed
Nov 04, 2026
Non-Final Rejection mailed — §101, §103 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12801984
SYSTEM AND METHOD FOR MONITORING AND REMOVING DRIFT IN MACHINE LEARNING MODELS
3y 7m to grant Granted Nov 17, 2026
Patent 12792908
METHOD FOR SOLVING PROBLEM AND SYSTEM THEREOF
3y 6m to grant Granted Nov 10, 2026
Patent 12783866
DATA AUGMENTATION FOR TRAINING ARTIFICIAL INTELLIGENCE MODEL
5y 0m to grant Granted Nov 03, 2026
Patent 12783864
COMPUTER-IMPLEMENTED METHOD FOR CREATING ENCODED DATA
4y 11m to grant Granted Nov 03, 2026
Patent 12770030
METHOD AND SYSTEM FOR EXCEPTION MANAGEMENT
3y 3m to grant Granted Oct 20, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

1-2
Expected OA Rounds
50%
Grant Probability
88%
With Interview (+38.1%)
3y 3m (~6m remaining)
Median Time to Grant
Low
PTA Risk
Based on 299 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month