CTNF 18/404,353 CTNF 101532 Detailed Action The following NON-FINAL office action is in response to application 18/404353 filed on 1/04/2024 . This communication is the first action on the merits. Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. 12-151 AIA 26-51 12-51 Status of Claims Claims 1-20 are currently pending and have been rejected as follows. Information Disclosure Statement The information disclosure statement (IDS) submitted on 6/24/2024 complies with the provisions of 37 CFR 1.97 and is being considered. Claim Rejections - 35 USC § 101 07-04-01 AIA 07-04 35 U.S.C. 101 reads as follows: Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title. Claims 1, 3-11, 13-17, and 19-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception without significantly more. A subject matter eligibility analysis is set forth below. See MPEP 2106. Specifically, representative Claim 1 recites: A method for identifying faults or fractures in a subsurface for performing hydrocarbon extraction , the method comprising: receiving a seismic image generated from a plurality of seismic traces, the seismic image representing features in the subsurface based on reflections of the plurality of seismic traces; accessing a data processing model, the data processing model trained to generate output values based on a prior distribution associated with one or more machine learning parameters and a posterior distribution associated with the one or more machine learning parameters; processing the seismic image using the data processing model, the processing including : sampling a value from the posterior distribution; updating the one or more machine learning parameters based on the sampled value; and generating a prediction value for one or more locations in the seismic image , the prediction value specifying a presence or absence of a fault or fracture at a location in the seismic image ; and based on the processing, generating an uncertainty value associated with the prediction value, the uncertainty value based on processing the seismic image with at least two values from the posterior distribution; and generating an output representation of the uncertainty value and the prediction value specifying the presence or the absence of the fault or the fracture at the location in the seismic image. The claim limitations in the abstract idea have been underlined above; the remaining limitations are “additional elements.” Similar limitations comprise the abstract idea of Claims 11 and 20. Step 1: Under Step 1 of the analysis, Claim 1 belongs to a statutory category, namely it is a method claim. Likewise, Claim 11 is a system claim and Claim 17 is a product claim. Step 2A – Prong I: Under Step 2A, prong 1: This part of the eligibility analysis evaluates whether the claim recites a judicial exception. As explained in MPEP 2106.04, subsection II, a claim “recites” a judicial exception when the judicial exception is “set forth” or “described” in the claim. In the instant case, Claim 1 is found to recite at least one judicial exception (i.e. abstract idea), that being a Mental Process and Mathematical Calculation. This can be seen in the following claim limitations: “ identifying faults or fractures in a subsurface for performing hydrocarbon extraction ,” “ processing the seismic image using the data processing model, the processing including ,” “ generating a prediction value for one or more locations in the seismic image, the prediction value specifying a presence or absence of a fault or fracture at a location in the seismic image ,” and “ based on the processing, generating an uncertainty value associated with the prediction value, the uncertainty value based on processing the seismic image. ” Subsurface fault identification in seismic images can be accomplished with the naked eye [ See Inst. Spec., Fig. [7a] – this is an example of a seismic image, where the discontinuities 702 and 704 can be detected with the human eye; From Paragraph [0095] – “Image 700 of FIG. 7A includes an example of a seismic image that is preprocessed as described in relation to FIGS. 1-4C. Discontinuities in the seismic image are shown at locations 702, 704. These discontinuities can represent faults or fractures.” ] and is merely a data observation and judgement. Processing the seismic image using the data processing model consists of a series of steps, two of which (generating a prediction value and generating an uncertainty value) have been identified as within the abstract idea, which will be elaborated below. Thus, the processing of the seismic image is considered a mathematical calculation and mental process as it consists of multiple steps that are also mathematical calculations and mental processes. Generating a prediction value is described [ See Inst. Spec., Paragraph [0075], Eq. [15] ] as one of the mathematical operations performed by the data processing model, generating the prediction value as the output given some machine learning parameter as the input. The uncertainty value is calculated from the mean and variance [ See Inst. Spec., Paragraph [0076] ] As such, the generating a prediction value and generating uncertainty value limitations are both mathematical calculations and mental processes. Similar limitations comprise the abstract ideas of Claim 11 and 17 . Step 2A – Prong II: Step 2A, prong 2 of the eligibility analysis evaluates whether the claim as a whole integrates the recited judicial exception(s) into a practical application of the exception. This evaluation is performed by (a) identifying whether there are any additional elements recited in the claim beyond the judicial exception, and (b) evaluating those additional elements individually and in combination to determine whether the claim as a whole integrates the exception into a practical application. The claims do not amount to the recitation of a particular practical application because they do not recite any specific steps that would improve the underlying hydrocarbon extraction process. Thus, under Step 2A, prong 2 of the analysis, even when viewed in combination, these additional elements do not integrate the recited judicial exception into a practical application and the claim is directed to the judicial exception. No specific practical application is associated with the claimed system (method and product for claims 11 and 17, respectively). Step 2B: Under Step 2B, the claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements, as described above with respect to Step 2A Prong 2, merely amount to a general purpose computer system that attempts to apply the abstract idea in a technological environment, limiting the abstract idea to a particular field of use, and merely perform insignificant extra-solution activit(ies). Claims 1, 11, and 17 all recite receiving a seismic image generated from seismic traces, which amounts to insignificant pre-solution activity, and generating an output representation of the uncertainty and prediction value, which is insignificant post-solution activity. Such insignificant extra-solution activity, e.g. data gathering and output, when re-evaluated under Step 2B is further found to be well-understood, routine, and conventional as evidenced by MPEP 2106.05(d)(II) (describing conventional activities that include transmitting and receiving data over a network, electronic recordkeeping, storing and retrieving information from memory, and electronically scanning or extracting data from a physical document)”. In addition to the abstract ideas recited in claims 1, 11, and 17, the claimed method recites the following additional elements: “ the data processing model trained to generate output values based on a prior distribution associated with one or more machine learning parameters and a posterior distribution associated with the one or more machine learning parameters ,” “ receiving a seismic image generated from a plurality of seismic traces, the seismic image representing features in the subsurface based on reflections of the plurality of seismic trace s,” “ accessing a data processing model ,” “ sampling a value from the posterior distribution ,” “ updating the one or more machine learning parameters based on the sampled value ,” “ with at least two values from the posterior distribution ,” and “ generating an output representation of the uncertainty value and the prediction value specifying the presence or the absence of the fault or the fracture at the location in the seismic image. ” Claim 11 recites no further additional elements while Claim 17 recites “One or more non-transitory computer readable media storing instructions…One or more non-transitory computer readable media storing instructions for identifying faults or fractures in a subsurface for performing hydrocarbon extraction, the instructions, when executed by at least one processor, configured to cause the at least one processor to perform operations. ” The data processing model is a Bayesian deep learning neural network [ See Inst. Spec. Paragraph [0033] ], which, as recited, amounts to no more than a general purpose machine learning algorithm that executes a series of functions and mathematical calculations that are well understood, routine, and conventional steps in a Bayesian Neural Network [ See Mosser et. al. (“Deep Bayesian neural networks for Fault identification and Uncertainty Quantification,” EAGE Digitalization Conference and Exhibition, November 30 – December 3, 2020, Vienna, Austria, 5 pages) – methodology section ; See Evans et. al. (US 11113632 B1) – Col. 7, Ln. [7-31] – “Bayesian inference…variational inference” ]. Receiving a seismic image is a data gathering step with no claimed recitation of how the images are received, and the fact that they are seismic images merely links the abovementioned mathematical calculations and mental processes to the field of seismic imaging. Accessing a data processing model is merely a data gathering/output step, as data or information must be transferred to access the data processing model, making this limitation represents no more than a step necessary to execute the data processing model. Sampling the posterior distribution and updating the machine learning parameters based on the sampled value are steps, or functions, of the data processing model and are well-understood, routine, and conventional steps in executing a Bayesian Neural Network [ See Ray et. al. (US 20190011583 A1), Paragraph [0009] – “sampling”, Paragraph [0051] – “updating”; See Evans et. al. – Col. 7, Ln. [7-31] – “sampling”, Col. 7, Ln. 47-53 – “updating” ]. The element “One or more non-transitory computer readable media storing instructions… when executed by at least one processor, configured to cause the at least one processor to perform operations” recites that the steps for identifying fractures are performed “by at least one processor” however this is found to be a general purpose computer and does not integrate the abstract idea into a practical application. See Alice Corp. v. CLS Bank International, 573 U.S. 208 (2014). The at least two values from the posterior distribution places a lower limit on the data to be sampled for use in the mathematical calculation, and as such, is a generic data gathering step. Generating an output representation of the uncertainty and prediction values specifying the presence or absence of the fault or fracture is a data output step visualizing the results of the abstract idea [ See Inst. Spec. Paragraph [0102], Fig. [9] ]. The abovementioned elements are found to be data gathering and output steps, which are recited at a high level of generality, and thus merely amount to “insignificant extra-solution” activity(ies) . See MPEP 2106.05(g) “Insignificant Extra-Solution Activity. Thus, these limitations also do not amount to significantly more than the judicial exception. See MPEP 2106.05(f). The additional elements listed above can also be viewed as attempts to generally link the abstract ideas of the claim to the environment of seismic imaging and fault/fracture detection. See MPEP 2106.05(h). The generic data gathering, processing, and output steps, are recited at such a high level of generality (that they represent no more than mere instructions to apply the judicial exceptions on a computer which can also be viewed as nothing more than an attempt to generally link the use of the judicial exceptions to the technological environment of a computer. Noting MPEP 2106.04(d)(I): “ It is notable that mere physicality or tangibility of an additional element or elements is not a relevant consideration in Step 2A Prong Two. As the Supreme Court explained in Alice Corp., mere physical or tangible implementation of an exception does not guarantee eligibility. Alice Corp. Pty. Ltd. v. CLS Bank Int’l, 573 U.S. 208, 224, 110 USPQ2d 1976, 1983-84 (2014) ("The fact that a computer ‘necessarily exist[s] in the physical, rather than purely conceptual, realm,’ is beside the point") ”. Therefore, similarly the combination and arrangement of the above identified additional elements when analyzed under Step 2B also fails to necessitate a conclusion that claims 1, 7, and 11 amount to significantly more than the abstract idea. With regards to the dependent claims, Claims 3-10, 13-16, and 19-20 merely further expand upon the algorithm/abstract idea and do not set forth further additional elements that integrate the recited abstract idea into a practical application or amount to significantly more. Therefore, these claims are found ineligible for the reasons described for parent claims 1, 11, and 17 Specifically: Claims 3 and 13 specify that the data processing model, which has already been identified as a general purpose computer not integrating the judicial exception into practical application or amounting to significantly more, comprises a Bayesian neural network (BNN). The BNN performs all the recited functions of the general purpose computer and is thus recited with such a level of generality that the abstract idea of the claims is not integrated into practical application, nor do the claims amount to significantly more than the abstract idea. Claims 4, 14, and 19 provide further limitations for the data processing model. Specifying that the training images are labeled is merely defining a data type to implement the training, while the training itself by “maximizing an evidence lower bound value based on a stochastic gradient ascent” is a routine step in implementing a Bayesian neural network [ See Evans et. al. (US 11113632 B1) , Col. 7, Ln. 7-31; Su et. al. (US 20240086716 A1), Paragraph [0049] ]. Thus, these claims are no more than generically recited functions of a general purpose computer and do not integrate the abstract idea into practical application or amount to significantly more than the judicial exceptions. Claims 5, 15, and 20 recite performing iterations of the data processing model until a threshold value for uncertainty is reached, which amounts to identifying and comparing model output with a numerical value and is included within the abstract idea. These limitations are parameters determined via mathematical calculations and mental processes and are within the abstract ideas of parent claims 1, 11, and 17 and thus do not provide any additional elements to integrate the claims into practical application or for to amount to significantly more than the judicial exceptions. Claims 6 and 16 recite random sampling, which is merely a data gathering step and as such, is an additional element, but does not integrate the claims into practical application or amount to significantly more than the judicial exception when viewed as a whole. Claim 7 defines the posterior distribution as a Gaussian. This merely further defines a mathematical function for a distribution and thus does not integrate the claimed method into practical application or amount to significantly more than the judicial exception when viewed as a whole. Claims 8 and 9 define parameters of the machine learning model as bias and weight, respectively. Bias and weight are common machine learning parameters used to modify the prior distribution in neural networks and thus are considered to be well-understood, conventional, and routine in the art [ See Iqbal et. al. (US 20250030435 A1), Paragraph [0048]; Colombo et. al. ( US 20230288592 A1 ), Paragraph [0040] ]. Thus, Claims 8 and 9 do not integrate the judicial exception into practical application or amount to significantly more than the recitation of components of a general purpose machine learning algorithm. Claim 10 specifies that the location in the seismic image is a pixel, which merely defines a data type for the output of the data processing model. While it is an additional element, it places a further limitation on the data output, which has been identified as insignificant post-solution activity. Thus, the claim is not integrated into practical application, nor does it amount to significantly more than the judicial exception when viewed as a whole. Claim Rejections - 35 USC § 103 07-06 AIA 15-10-15 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. 07-20-aia AIA The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. Claims 1,3,5-7,10 11,13,15-17, and 20 are rejected under Zhang et. al (US 20200292723 A1) in view of Zheng et. al (US 20230251395 A1). Regarding Claim 1 , Zhang discloses a method for identifying faults or fractures in a subsurface for performing hydrocarbon extraction, the method comprising: receiving a seismic image generated from a plurality of seismic traces [ Paragraph [0022] – “Referring now to FIG. 1, at block 12 , locations and properties of hydrocarbon deposits within a subsurface region of the Earth associated with the respective seismic survey may be determined based on the analyzed seismic data. In one embodiment, the seismic data acquired may be analyzed to generate a map or profile that illustrates various geological formations within the subsurface region . Based on the identified locations and properties of the hydrocarbon deposits, at block… h ydrocarbon exploration organizations may use the locations and properties of the hydrocarbon deposits and the associated overburdens to determine a path along which to drill into the Earth. ”; Paragraph [0026] – “The marine survey system 22 may include a vessel 30 , one or more seismic sources 32 , a (seismic) streamer 34 , one or more (seismic) receivers 36 , and/or other equipment that may assist in acquiring seismic images representative of geological formations within a subsurface region 26 of the Earth. The vessel 30 may tow the seismic source(s) 32 (e.g., an air gun array) that may produce energy, such as sound waves (e.g., seismic waveforms), that is directed at a seafloor 28 . The vessel 30 may also tow the streamer 34 having a receiver 36 (e.g., hydrophones) that may acquire seismic waveforms that represent the energy output by the seismic source(s) 32 subsequent to being reflected off of various geological formations (e.g., salt domes, faults , folds, etc.) within the subsurface region 26 .” ], the seismic image representing features in the subsurface based on reflections of the plurality of seismic traces [ Paragraph [0026] – “The marine survey system 22 may include a vessel 30 , one or more seismic sources 32 , a (seismic) streamer 34 , one or more (seismic) receivers 36 , and/or other equipment that may assist in acquiring seismic images representative of geological formations within a subsurface region 26 of the Earth. The vessel 30 may tow the seismic source(s) 32 (e.g., an air gun array) that may produce energy, such as sound waves (e.g., seismic waveforms), that is directed at a seafloor 28 . The vessel 30 may also tow the streamer 34 having a receiver 36 (e.g., hydrophones) that may acquire seismic waveforms that represent the energy output by the seismic source(s) 32 subsequent to being reflected off of various geological formations (e.g., salt domes, faults , folds, etc.) within the subsurface region 26 .” ]. Zhang does not disclose accessing a data processing model, the data processing model trained to generate output values based on a prior distribution associated with one or more machine learning parameters and a posterior distribution associated with the one or more machine learning parameters; processing the seismic image using the data processing model, the processing including: sampling a value from the posterior distribution or updating the one or more machine learning parameters based on the sampled value. However, Zheng discloses accessing a data processing model, the data processing model trained to generate output values [ Paragraph [0042] – “FIG. 5 illustrates one example of a technique, illustrated as a flow chart 78 , to carry out Bayesian inversion on processed seismic data in order to estimate subsurface elastic parameters and their associated uncertainties . This process can be performed on the computing system 60 to analyze acquired seismic data (e.g., performed as code stored on a tangible and non-transitory machine readable medium, such as the memory 66 and/or the storage 68 , that when in operation causes the processor 64 to perform one or more of the steps of the flow chart 78 as performance of the technique) …In step 82 geological and petrophysical data (i.e., seismic data) is received by computing system 60 . Examples of this data may be well logs, geological descriptions (rock types, etc), and geophysical data. In step 84 , the computing system 60 operates to train prior models.” – Bayesian inversion is the data processing model; model is accessed when computing system 60 runs process 78 (refer to Figs. [4] and [5]); subsurface elastic parameters and their associated uncertainties are the output ] based on a prior distribution associated with one or more machine learning parameters and a posterior distribution associated with the one or more machine learning parameters [ Paragraph [0043] – “ Probability distributions utilized in conjunction with the technique illustrated in FIG. 5 can be represented by a set of earth models, parameterized in terms of P-wave speed (Vp), S-wave speed (Vs), and density (e.g., density characteristics of a rock formation or type), and/or other physical representations of the subsurface which may be referred to collectively as particles . In some embodiments, the training of the prior model in step 84 includes the use of Gaussian, uniform, and/or Gaussian mixture models to generate statistical models as the prior model (or portions thereof) .” – note that probability distributions are being referred to as ‘particles’ and prior model is comprised of ‘particles’; Paragraph [0049] – “As part of process 100 , at each iteration, prior probabilities are combined with the likelihoods of each particle to form the target density and its gradient is weighted by a kernel function to provide the update directions for each particle. This process is then repeated until it arrives at a final set of particles that closely approximates the posterior distribution .” – prior probabilities are the prior distribution, kernel function is a machine learning parameter ]; processing the seismic image using the data processing model [ Paragraph [0037] – “… the computing system 60 may be employed to analyze the acquired seismic data to obtain an image representative of the subsurface region …” – processing the seismic image; Paragraph [0042] – “F IG. 5 illustrates one example of a technique, illustrated as a flow chart 78 , to carry out Bayesian inversion on processed seismic data in order to estimate subsurface elastic parameters and their associated uncertainties.” – Bayesian inversion is the data processing model, which is being employed by the computing system 60 ], the processing including: sampling a value from the posterior distribution [ Paragraph [0041] – “In some embodiments, Bayesian inference is undertaken (e.g., Bayesian inference applied to the seismic data) using a probabilistic approach utilizing a sampling method, such as Markov Chain Monte Carlo (MCMC), to draw samples from the posterior distribution. ” ]; and updating the one or more machine learning parameters based on the sampled value [ Paragraph [0049] – “ As part of process 100 , at each iteration, prior probabilities are combined with the likelihoods of each particle to form the target density and its gradient is weighted by a kernel function to provide the update directions for each particle . This process is then repeated until it arrives at a final set of particles that closely approximates the posterior distribution .” – particle updates are made based on similarity with the sampled posterior distribution; the kernel function is the machine learning parameter; Paragraph [0050] – “Step 106 of process 100 includes evaluation of the kernel function that is used to scale the above described gradient scores …The kernel function is a mathematical measure of similarity between the particles, but in the context of the algorithm it enforces diversity in the posterior. The computed gradients of step 108 are then used to update all the particles simultaneously in each iteration in step 110 . At step 112 , a determination on whether convergence has occurred is undertaken … If the determination is negative with respect to the convergence in step 112 , process 98 is undertaken with the revised values for the particles. ” – See Fig. [5], steps 98-112 are repeated with updated particles until convergence, updating particles updates the kernel function, which is the machine learning parameter ]. It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to apply the data processing model of Zheng to the fault and fracture detection method of Zhang in order to improve the prediction of fault and fracture locations in seismic images. The combination of Zhang and Zheng would disclose generating a prediction value for one or more locations in the seismic image, the prediction value specifying a presence or absence of a fault or fracture at a location in the seismic image [Zhang, Paragraph [0049] – “Image 88 includes center point 90. Fault prediction can be treated as an image classification problem , whereby the neural networks 92 and 94 classify only a particular location (e.g., the center point 90) of an image/cube (e.g., image 88) as indicative of a fault or not. ” – see Fig. [6]; Paragraph [0050] – “… neural network 92 can predict and generate as an output the dip (e.g., the angle of a fault relative to a horizontal plane) of a fault located at or about the center point 90 … can predict and generate as an output the azimuth (e.g., the angle characterizing direction of the fault with respect to a reference direction) of a fault located at or about the center point 90 as a portion of step 82 of FIG. 5.” – prediction values, see Fig. [5] ]; and based on the processing, generating an uncertainty value associated with the prediction value, the uncertainty value based on processing the seismic image with at least two values from the posterior distribution [ Zheng, Paragraph [0037] – “As such, the computing system 60 may be employed to analyze the acquired seismic data to obtain an image representative of the subsurface region …” – Bayesian inversion technique is being applied to seismic images; Paragraph [0041] – “In some embodiments, Bayesian inference is undertaken (e.g., Bayesian inference applied to the seismic data) using a probabilistic approach utilizing a sampling method, such as Markov Chain Monte Carlo (MCMC), to draw samples from the posterior distribution.” – multiple samples; Paragraph [0052] – “The above described technique to carry out Bayesian inversion on processed seismic data in order to estimate subsurface elastic parameters and their associated uncertaint ies provides benefits…” ]; generating an output representation of the uncertainty value and the prediction value [Zheng, Paragraph [0051] – “FIG. 7 illustrates a graphical representation 116 of estimates of the posterior distribution 118 as well as a mean of the posterior for a volume 120 in a 3D volume. Accordingly, FIG. 7 represents an output generated using the AVO inversion performed using a probabilistic approach (e.g., Bayesian approach) as discussed above with respect to FIG. 5 .” – see Fig. [7], 120 ] specifying the presence or the absence of the fault or the fracture at the location in the seismic image [Zhang, Paragraph [0052]-[0053] – “In step 96 , if either or both of the indications received as outputs from the neural networks 92 and 94 are negative indications , in step 96 , the computing system 60 determines that no fault is present in image 88 (at the center point) , as a portion of step 84 of FIG. 5, and the computing system 60 generates an output 98 indicating (e.g., classifying) image 88 as having no fault (at the center point). However, if the output from both of the neural network 92 and the neural network 94 indicate the presence of a fault (at the center point) , the computing system 60 determines that a fault is present in image 88 , as a portion of step 84 of FIG. 5, and the computing system 60 generates an output 98 indicating (e.g., classifying) image 88 as having a fault (with the respective aspects, such as dip and azimuth, corresponding to the fault). Thus, a center point 90 of an image 88 is determined to be a fault (or have a fault therein) …” ]. Regarding Claim 11, Zhang discloses a system for identifying faults or fractures in a subsurface for performing hydrocarbon extraction [ Paragraph [0022] – “Referring now to FIG. 1, at block 12 , locations and properties of hydrocarbon deposits within a subsurface region of the Earth associated with the respective seismic survey may be determined based on the analyzed seismic data. In one embodiment, the seismic data acquired may be analyzed to generate a map or profile that illustrates various geological formations within the subsurface region . Based on the identified locations and properties of the hydrocarbon deposits, at block… h ydrocarbon exploration organizations may use the locations and properties of the hydrocarbon deposits and the associated overburdens to determine a path along which to drill into the Earth. ”; Paragraph [0026] – “The marine survey system 22 may include a vessel 30 , one or more seismic sources 32 , a (seismic) streamer 34 , one or more (seismic) receivers 36 , and/or other equipment that may assist in acquiring seismic images representative of geological formations within a subsurface region 26 of the Earth. The vessel 30 may tow the seismic source(s) 32 (e.g., an air gun array) that may produce energy, such as sound waves (e.g., seismic waveforms), that is directed at a seafloor 28 . The vessel 30 may also tow the streamer 34 having a receiver 36 (e.g., hydrophones) that may acquire seismic waveforms that represent the energy output by the seismic source(s) 32 subsequent to being reflected off of various geological formations (e.g., salt domes, faults , folds, etc.) within the subsurface region 26 .” ], the system comprising: at least one processor [ See Fig. [4], processor in block 64 ]; and a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations comprising [ Paragraph [0031] – “ The processor 64 may be any type of computer processor or microprocessor capable of executing computer-executable code or instructions to implement the methods described herein . The processor 64 may also include multiple processors that may perform the operations described below. The memory 66 and the storage 68 may be any suitable articles of manufacture that can serve as media to store processor-executable code, data, or the like . These articles of manufacture may represent computer-readable media (e.g., any suitable form of memory or storage) that may store the processor-executable code used by the processor 64 to perform the presently disclosed techniques. ” ]: receiving a seismic image generated from a plurality of seismic traces, the seismic image representing features in the subsurface based on reflections of the plurality of seismic traces. Zhang does not disclose accessing a data processing model, the data processing model trained to generate output values based on a prior distribution associated with one or more machine learning parameters and a posterior distribution associated with the one or more machine learning parameters; processing the seismic image using the data processing model, the processing including: sampling a value from the posterior distribution or updating the one or more machine learning parameters based on the sampled value. However, Zheng discloses accessing a data processing model, the data processing model trained to generate output values [ Paragraph [0042] – “FIG. 5 illustrates one example of a technique, illustrated as a flow chart 78 , to carry out Bayesian inversion on processed seismic data in order to estimate subsurface elastic parameters and their associated uncertainties . This process can be performed on the computing system 60 to analyze acquired seismic data (e.g., performed as code stored on a tangible and non-transitory machine readable medium, such as the memory 66 and/or the storage 68 , that when in operation causes the processor 64 to perform one or more of the steps of the flow chart 78 as performance of the technique) …In step 82 geological and petrophysical data (i.e., seismic data) is received by computing system 60 . Examples of this data may be well logs, geological descriptions (rock types, etc), and geophysical data. In step 84 , the computing system 60 operates to train prior models.” – Bayesian inversion is the data processing model; model is accessed when computing system 60 runs process 78 (refer to Figs. [4] and [5]); subsurface elastic parameters and their associated uncertainties are the output ] based on a prior distribution associated with one or more machine learning parameters and a posterior distribution associated with the one or more machine learning parameters [ Paragraph [0043] – “ Probability distributions utilized in conjunction with the technique illustrated in FIG. 5 can be represented by a set of earth models, parameterized in terms of P-wave speed (Vp), S-wave speed (Vs), and density (e.g., density characteristics of a rock formation or type), and/or other physical representations of the subsurface which may be referred to collectively as particles . In some embodiments, the training of the prior model in step 84 includes the use of Gaussian, uniform, and/or Gaussian mixture models to generate statistical models as the prior model (or portions thereof) .” – note that probability distributions are being referred to as ‘particles’ and prior model is comprised of ‘particles’; Paragraph [0049] – “As part of process 100 , at each iteration, prior probabilities are combined with the likelihoods of each particle to form the target density and its gradient is weighted by a kernel function to provide the update directions for each particle. This process is then repeated until it arrives at a final set of particles that closely approximates the posterior distribution .” – prior probabilities are the prior distribution, kernel function is a machine learning parameter ]; processing the seismic image using the data processing model [ Paragraph [0037] – “… the computing system 60 may be employed to analyze the acquired seismic data to obtain an image representative of the subsurface region …” – processing the seismic image; Paragraph [0042] – “F IG. 5 illustrates one example of a technique, illustrated as a flow chart 78 , to carry out Bayesian inversion on processed seismic data in order to estimate subsurface elastic parameters and their associated uncertainties.” – Bayesian inversion is the data processing model, which is being employed by the computing system 60 ], the processing including: sampling a value from the posterior distribution [ Paragraph [0041] – “In some embodiments, Bayesian inference is undertaken (e.g., Bayesian inference applied to the seismic data) using a probabilistic approach utilizing a sampling method, such as Markov Chain Monte Carlo (MCMC), to draw samples from the posterior distribution. ” ]; and updating the one or more machine learning parameters based on the sampled value [ Paragraph [0049] – “ As part of process 100 , at each iteration, prior probabilities are combined with the likelihoods of each particle to form the target density and its gradient is weighted by a kernel function to provide the update directions for each particle . This process is then repeated until it arrives at a final set of particles that closely approximates the posterior distribution .” – particle updates are made based on similarity with the sampled posterior distribution; the kernel function is the machine learning parameter; Paragraph [0050] – “Step 106 of process 100 includes evaluation of the kernel function that is used to scale the above described gradient scores …The kernel function is a mathematical measure of similarity between the particles, but in the context of the algorithm it enforces diversity in the posterior. The computed gradients of step 108 are then used to update all the particles simultaneously in each iteration in step 110 . At step 112 , a determination on whether convergence has occurred is undertaken … If the determination is negative with respect to the convergence in step 112 , process 98 is undertaken with the revised values for the particles. ” – See Fig. [5], steps 98-112 are repeated with updated particles until convergence, updating particles updates the kernel function, which is the machine learning parameter ]. It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to apply the data processing model of Zheng to the fault and fracture detection method of Zhang in order to improve the prediction of fault and fracture locations in seismic images. The combination of Zhang and Zheng would disclose generating a prediction value for one or more locations in the seismic image, the prediction value specifying a presence or absence of a fault or fracture at a location in the seismic image [Zhang, Paragraph [0049] – “Image 88 includes center point 90. Fault prediction can be treated as an image classification problem , whereby the neural networks 92 and 94 classify only a particular location (e.g., the center point 90) of an image/cube (e.g., image 88) as indicative of a fault or not. ” – see Fig. [6]; Paragraph [0050] – “… neural network 92 can predict and generate as an output the dip (e.g., the angle of a fault relative to a horizontal plane) of a fault located at or about the center point 90 … can predict and generate as an output the azimuth (e.g., the angle characterizing direction of the fault with respect to a reference direction) of a fault located at or about the center point 90 as a portion of step 82 of FIG. 5.” – prediction values, see Fig. [5] ]; and based on the processing, generating an uncertainty value associated with the prediction value, the uncertainty value based on processing the seismic image with at least two values from the posterior distribution [ Zheng, Paragraph [0037] – “As such, the computing system 60 may be employed to analyze the acquired seismic data to obtain an image representative of the subsurface region …” – Bayesian inversion technique is being applied to seismic images; Paragraph [0041] – “In some embodiments, Bayesian inference is undertaken (e.g., Bayesian inference applied to the seismic data) using a probabilistic approach utilizing a sampling method, such as Markov Chain Monte Carlo (MCMC), to draw samples from the posterior distribution.” – multiple samples; Paragraph [0052] – “The above described technique to carry out Bayesian inversion on processed seismic data in order to estimate subsurface elastic parameters and their associated uncertaint ies provides benefits…” ]; generating an output representation of the uncertainty value and the prediction value [Zheng, Paragraph [0051] – “FIG. 7 illustrates a graphical representation 116 of estimates of the posterior distribution 118 as well as a mean of the posterior for a volume 120 in a 3D volume. Accordingly, FIG. 7 represents an output generated using the AVO inversion performed using a probabilistic approach (e.g., Bayesian approach) as discussed above with respect to FIG. 5 .” – see Fig. [7], 120 ] specifying the presence or the absence of the fault or the fracture at the location in the seismic image [Zhang, Paragraph [0052]-[0053] – “In step 96 , if either or both of the indications received as outputs from the neural networks 92 and 94 are negative indications , in step 96 , the computing system 60 determines that no fault is present in image 88 (at the center point) , as a portion of step 84 of FIG. 5, and the computing system 60 generates an output 98 indicating (e.g., classifying) image 88 as having no fault (at the center point). However, if the output from both of the neural network 92 and the neural network 94 indicate the presence of a fault (at the center point) , the computing system 60 determines that a fault is present in image 88 , as a portion of step 84 of FIG. 5, and the computing system 60 generates an output 98 indicating (e.g., classifying) image 88 as having a fault (with the respective aspects, such as dip and azimuth, corresponding to the fault). Thus, a center point 90 of an image 88 is determined to be a fault (or have a fault therein) …” ]. Regarding Claim 17, Zhang discloses one or more non-transitory computer readable media storing instructions for identifying faults or fractures in a subsurface for performing hydrocarbon extraction [ Paragraph [0022] – “Referring now to FIG. 1, at block 12 , locations and properties of hydrocarbon deposits within a subsurface region of the Earth associated with the respective seismic survey may be determined based on the analyzed seismic data. In one embodiment, the seismic data acquired may be analyzed to generate a map or profile that illustrates various geological formations within the subsurface region . Based on the identified locations and properties of the hydrocarbon deposits, at block… hydrocarbon exploration organizations may use the locations and properties of the hydrocarbon deposits and the associated overburdens to determine a path along which to drill into the Earth. ”; Paragraph [0026] – “The marine survey system 22 may include a vessel 30 , one or more seismic sources 32 , a (seismic) streamer 34 , one or more (seismic) receivers 36 , and/or other equipment that may assist in acquiring seismic images representative of geological formations within a subsurface region 26 of the Earth. The vessel 30 may tow the seismic source(s) 32 (e.g., an air gun array) that may produce energy, such as sound waves (e.g., seismic waveforms), that is directed at a seafloor 28 . The vessel 30 may also tow the streamer 34 having a receiver 36 (e.g., hydrophones) that may acquire seismic waveforms that represent the energy output by the seismic source(s) 32 subsequent to being reflected off of various geological formations (e.g., salt domes, faults , folds, etc.) within the subsurface region 26 .”; Paragraph [0031] – “The processor 64 may also include multiple processors that may perform the operations described below. The memory 66 and the storage 68 may be any suitable articles of manufacture that can serve as media to store processor-executable code, data, or the like. These articles of manufacture may represent computer-readable media (e.g., any suitable form of memory or storage) that may store the processor-executable code used by the processor 64 to perform the presently disclosed techniques. ” ], the instructions, when executed by at least one processor, configured to cause the at least one processor to perform operations comprising [ Paragraph [0031] – “ These articles of manufacture may represent computer-readable media (e.g., any suitable form of memory or storage) that may store the processor-executable code used by the processor 64 to perform the presently disclosed techniques. ” ]: receiving a seismic image generated from a plurality of seismic traces, the seismic image representing features in the subsurface based on reflections of the plurality of seismic traces. Zhang does not disclose accessing a data processing model, the data processing model trained to generate output values based on a prior distribution associated with one or more machine learning parameters and a posterior distribution associated with the one or more machine learning parameters; processing the seismic image using the data processing model, the processing including: sampling a value from the posterior distribution or updating the one or more machine learning parameters based on the sampled value. However, Zheng discloses accessing a data processing model, the data processing model trained to generate output values [ Paragraph [0042] – “FIG. 5 illustrates one example of a technique, illustrated as a flow chart 78 , to carry out Bayesian inversion on processed seismic data in order to estimate subsurface elastic parameters and their associated uncertainties . This process can be performed on the computing system 60 to analyze acquired seismic data (e.g., performed as code stored on a tangible and non-transitory machine readable medium, such as the memory 66 and/or the storage 68 , that when in operation causes the processor 64 to perform one or more of the steps of the flow chart 78 as performance of the technique) …In step 82 geological and petrophysical data (i.e., seismic data) is received by computing system 60 . Examples of this data may be well logs, geological descriptions (rock types, etc), and geophysical data. In step 84 , the computing system 60 operates to train prior models.” – Bayesian inversion is the data processing model; model is accessed when computing system 60 runs process 78 (refer to Figs. [4] and [5]); subsurface elastic parameters and their associated uncertainties are the output ] based on a prior distribution associated with one or more machine learning parameters and a posterior distribution associated with the one or more machine learning parameters [ Paragraph [0043] – “ Probability distributions utilized in conjunction with the technique illustrated in FIG. 5 can be represented by a set of earth models, parameterized in terms of P-wave speed (Vp), S-wave speed (Vs), and density (e.g., density characteristics of a rock formation or type), and/or other physical representations of the subsurface which may be referred to collectively as particles . In some embodiments, the training of the prior model in step 84 includes the use of Gaussian, uniform, and/or Gaussian mixture models to generate statistical models as the prior model (or portions thereof) .” – note that probability distributions are being referred to as ‘particles’ and prior model is comprised of ‘particles’; Paragraph [0049] – “As part of process 100 , at each iteration, prior probabilities are combined with the likelihoods of each particle to form the target density and its gradient is weighted by a kernel function to provide the update directions for each particle. This process is then repeated until it arrives at a final set of particles that closely approximates the posterior distribution .” – prior probabilities are the prior distribution, kernel function is a machine learning parameter ]; processing the seismic image using the data processing model [ Paragraph [0037] – “… the computing system 60 may be employed to analyze the acquired seismic data to obtain an image representative of the subsurface region …” – processing the seismic image; Paragraph [0042] – “F IG. 5 illustrates one example of a technique, illustrated as a flow chart 78 , to carry out Bayesian inversion on processed seismic data in order to estimate subsurface elastic parameters and their associated uncertainties.” – Bayesian inversion is the data processing model, which is being employed by the computing system 60 ], the processing including: sampling a value from the posterior distribution [ Paragraph [0041] – “In some embodiments, Bayesian inference is undertaken (e.g., Bayesian inference applied to the seismic data) using a probabilistic approach utilizing a sampling method, such as Markov Chain Monte Carlo (MCMC), to draw samples from the posterior distribution. ” ]; and updating the one or more machine learning parameters based on the sampled value [ Paragraph [0049] – “ As part of process 100 , at each iteration, prior probabilities are combined with the likelihoods of each particle to form the target density and its gradient is weighted by a kernel function to provide the update directions for each particle . This process is then repeated until it arrives at a final set of particles that closely approximates the posterior distribution .” – particle updates are made based on similarity with the sampled posterior distribution; the kernel function is the machine learning parameter; Paragraph [0050] – “Step 106 of process 100 includes evaluation of the kernel function that is used to scale the above described gradient scores …The kernel function is a mathematical measure of similarity between the particles, but in the context of the algorithm it enforces diversity in the posterior. The computed gradients of step 108 are then used to update all the particles simultaneously in each iteration in step 110 . At step 112 , a determination on whether convergence has occurred is undertaken … If the determination is negative with respect to the convergence in step 112 , process 98 is undertaken with the revised values for the particles. ” – See Fig. [5], steps 98-112 are repeated with updated particles until convergence, updating particles updates the kernel function, which is the machine learning parameter ]. It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to apply the data processing model of Zheng to the fault and fracture detection method of Zhang in order to improve the prediction of fault and fracture locations in seismic images. The combination of Zhang and Zheng would disclose generating a prediction value for one or more locations in the seismic image, the prediction value specifying a presence or absence of a fault or fracture at a location in the seismic image [Zhang, Paragraph [0049] – “Image 88 includes center point 90. Fault prediction can be treated as an image classification problem , whereby the neural networks 92 and 94 classify only a particular location (e.g., the center point 90) of an image/cube (e.g., image 88) as indicative of a fault or not. ” – see Fig. [6]; Paragraph [0050] – “… neural network 92 can predict and generate as an output the dip (e.g., the angle of a fault relative to a horizontal plane) of a fault located at or about the center point 90 … can predict and generate as an output the azimuth (e.g., the angle characterizing direction of the fault with respect to a reference direction) of a fault located at or about the center point 90 as a portion of step 82 of FIG. 5.” – prediction values, see Fig. [5] ]; and based on the processing, generating an uncertainty value associated with the prediction value, the uncertainty value based on processing the seismic image with at least two values from the posterior distribution [ Zheng, Paragraph [0037] – “As such, the computing system 60 may be employed to analyze the acquired seismic data to obtain an image representative of the subsurface region …” – Bayesian inversion technique is being applied to seismic images; Paragraph [0041] – “In some embodiments, Bayesian inference is undertaken (e.g., Bayesian inference applied to the seismic data) using a probabilistic approach utilizing a sampling method, such as Markov Chain Monte Carlo (MCMC), to draw samples from the posterior distribution.” – multiple samples; Paragraph [0052] – “The above described technique to carry out Bayesian inversion on processed seismic data in order to estimate subsurface elastic parameters and their associated uncertaint ies provides benefits…” ]; generating an output representation of the uncertainty value and the prediction value [Zheng, Paragraph [0051] – “FIG. 7 illustrates a graphical representation 116 of estimates of the posterior distribution 118 as well as a mean of the posterior for a volume 120 in a 3D volume. Accordingly, FIG. 7 represents an output generated using the AVO inversion performed using a probabilistic approach (e.g., Bayesian approach) as discussed above with respect to FIG. 5 .” – see Fig. [7], 120 ] specifying the presence or the absence of the fault or the fracture at the location in the seismic image [Zhang, Paragraph [0052]-[0053] – “In step 96 , if either or both of the indications received as outputs from the neural networks 92 and 94 are negative indications , in step 96 , the computing system 60 determines that no fault is present in image 88 (at the center point) , as a portion of step 84 of FIG. 5, and the computing system 60 generates an output 98 indicating (e.g., classifying) image 88 as having no fault (at the center point). However, if the output from both of the neural network 92 and the neural network 94 indicate the presence of a fault (at the center point) , the computing system 60 determines that a fault is present in image 88 , as a portion of step 84 of FIG. 5, and the computing system 60 generates an output 98 indicating (e.g., classifying) image 88 as having a fault (with the respective aspects, such as dip and azimuth, corresponding to the fault). Thus, a center point 90 of an image 88 is determined to be a fault (or have a fault therein) …” ]. Regarding Claim 3, Zhang and Zheng disclose the method of claim 1, wherein the data processing model comprises a Bayesian neural network [ Zheng , Paragraph [0028] – “With one or more embodiments, processor 64 can instantiate or operate in conjunction with one or more seismic inversion techniques. With another embodiment, the computing system 60 can be implemented by using neural networks. ”; Paragraph [0047] – “Returning to FIG. 5 , flow chart 78 additionally includes process 98 , which represents forward modeling, and process 100 , which represents the computing system 60 running an efficient particle-based inference algorithm known as Stein Variational Gradient Descent (SVGD). To perform Bayesian optimization (e.g., a Bayesian inversion) with SVGD, probability distributions are represented by sets of particles instead of probability density functions. ” – computing system implements the Bayesian optimization and can do so via neural networks ]. Regarding Claim 13, Zhang and Zheng disclose the system of claim 11, wherein the data processing model comprises a Bayesian neural network [ Zheng , Paragraph [0028] – “With one or more embodiments, processor 64 can instantiate or operate in conjunction with one or more seismic inversion techniques. With another embodiment, the computing system 60 can be implemented by using neural networks. ”; Paragraph [0047] – “Returning to FIG. 5 , flow chart 78 additionally includes process 98 , which represents forward modeling, and process 100 , which represents the computing system 60 running an efficient particle-based inference algorithm known as Stein Variational Gradient Descent (SVGD). To perform Bayesian optimization (e.g., a Bayesian inversion) with SVGD, probability distributions are represented by sets of particles instead of probability density functions .” – computing system implements the Bayesian optimization and can do so via neural networks ]. Regarding Claim 5, Zhang and Zheng would disclose the method of claim 1, further comprising performing iterations of the processing the seismic image using the data processing model until a threshold value is satisfied for the uncertainty for at least one location in the seismic image [ Zheng, Paragraph [0037] “As such, the computing system 60 may be employed to analyze the acquired seismic data to obtain an image representative of the subsurface region …” – data processing model is processing the seismic images; Paragraph [0049] – “As part of process 100 , at each iteration , prior probabilities are combined with the likelihoods of each particle to form the target density and its gradient is weighted by a kernel function to provide the update directions for each particle. This process is then repeated until it arrives at a final set of particles that closely approximates the posterior distribution ...After a predetermined number of iterations and/or when a pre-determined convergence criterion is met, a final set of particles that represents the target (posterior) distribution is generated.” – pre-determined convergence criterion is a threshold; See Fig. [7] for model output depicting fault location prediction and uncertainty in seismic image ]. Regarding Claim 15, Zhang and Zheng would disclose the system of claim 11, the operations further comprising performing iterations of the processing the seismic image using the data processing model until a threshold value is satisfied for the uncertainty for at least one location in the seismic image [ Zheng, Paragraph [0037] “As such, the computing system 60 may be employed to analyze the acquired seismic data to obtain an image representative of the subsurface region …” – data processing model is processing the seismic images; Paragraph [0049] – “As part of process 100 , at each iteration , prior probabilities are combined with the likelihoods of each particle to form the target density and its gradient is weighted by a kernel function to provide the update directions for each particle. This process is then repeated until it arrives at a final set of particles that closely approximates the posterior distribution ...After a predetermined number of iterations and/or when a pre-determined convergence criterion is met, a final set of particles that represents the target (posterior) distribution is generated.” – pre-determined convergence criterion is a threshold; See Fig. [7] for model output depicting fault location prediction and uncertainty in seismic image ]. Regarding Claim 20, Zhang and Zheng would disclose the one or more non-transitory computer readable media of claim 17, the operations further comprising performing iterations of the processing the seismic image using the data processing model until a threshold value is satisfied for the uncertainty for at least one location in the seismic image [ Zheng, Paragraph [0037] “As such, the computing system 60 may be employed to analyze the acquired seismic data to obtain an image representative of the subsurface region …” – data processing model is processing the seismic images; Paragraph [0049] – “As part of process 100 , at each iteration , prior probabilities are combined with the likelihoods of each particle to form the target density and its gradient is weighted by a kernel function to provide the update directions for each particle. This process is then repeated until it arrives at a final set of particles that closely approximates the posterior distribution ...After a predetermined number of iterations and/or when a pre-determined convergence criterion is met, a final set of particles that represents the target (posterior) distribution is generated.” – pre-determined convergence criterion is a threshold; See Fig. [7] for model output depicting fault location prediction and uncertainty in seismic image ]. Regarding Claim 6, Zhang and Zheng would disclose the method of claim 1, wherein sampling comprises a random sampling across the posterior distribution [ Zheng, Paragraph [0041] – “… utilizing a sampling method, such as Markov Chain Monte Carlo (MCMC), to draw samples from the posterior distribution. ” – MCMC is a random sampling technique]. Regarding Claim 16, Zhang and Zheng would disclose the system of claim 11, wherein sampling comprises a random sampling across the posterior distribution [ Zheng, Paragraph [0041] – “… utilizing a sampling method, such as Markov Chain Monte Carlo (MCMC), to draw samples from the posterior distribution. ” – MCMC is a random sampling technique]. Regarding Claim 7, Zhang and Zheng would disclose method of claim 1, wherein the distribution comprises a Gaussian distribution [ Zheng, Paragraph [0043] – “In some embodiments, the training of the prior model in step 84 includes the use of Gaussian, uniform, and/or Gaussian mixture models to generate statistical models as the prior model (or portions thereof). ” ]. Regarding Claim 10, Zhang and Zheng would disclose method of claim 1, wherein the location in the seismic image comprises a pixel in the seismic image [ Paragraph [0049] – “ Fault prediction can be treated as an image classification problem, whereby the neural networks 92 and 94 classify only a particular location (e.g., the center point 90 ) of an image/cube (e.g., image 88 ) as indicative of a fault or not. When predicting faults using this technique, (e.g., a center point classifier), a sliding window is moved across a whole of the seismic image to be processed, typically voxel by voxel (or pixel by pixel) .” ]. Claims 2, 12, and 18 are rejected under Zhang et. al in view of Zheng et. al in further view of Rueger et. al. (US 11307319 B2). Regarding Claim 2 , the combination of Zhang and Zheng would disclose the method of claim 1. Zhang does not disclose based on the prediction value and the uncertainty value, generating a control signal configured for causing drilling of a well in the subsurface corresponding to the location in the seismic image. Rueger, however, discloses based on the prediction value and the uncertainty value, generating a control signal configured for causing drilling of a well in the subsurface corresponding to the location in the seismic image [ Col. 8, Ln. 29-39 - “For example, the computing device 600 may execute the action module 618 to control a drilling operation within a subterranean formation . In such an example, a drilling direction may be determined based on a predicted location of a fault in the subterranean formation prior to drilling into the fault. The control of the drilling operation may also generally include making a determination about whether to drill a well based on economic risk implied by the volumetric uncertainty provided by the fault uncertainty confidence indexes 412 . ”; Col. 8 Ln. 56-59 – “The computing device 600 can generate and display a GUI that includes an alert indicating whether the particular areas are suitable for further exploration or should be avoided during a drilling operation. ” ]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to generate the control signal/alert of Rueger following the subsurface fault location detection of Zheng and Zhang in order to signify locations to drill into the Earth. Regarding Claim 12, the combination of Zhang and Zheng would disclose the system of claim 11. The combination does not disclose the operations further comprising: based on the prediction value and the uncertainty value, generating a control signal configured for causing drilling of a well in the subsurface corresponding to the location in the seismic image. However, Rueger discloses the operations further comprising: based on the prediction value and the uncertainty value, generating a control signal configured for causing drilling of a well in the subsurface corresponding to the location in the seismic image [ Col. 8, Ln. 29-39 - “For example, the computing device 600 may execute the action module 618 to control a drilling operation within a subterranean formation . In such an example, a drilling direction may be determined based on a predicted location of a fault in the subterranean formation prior to drilling into the fault. The control of the drilling operation may also generally include making a determination about whether to drill a well based on economic risk implied by the volumetric uncertainty provided by the fault uncertainty confidence indexes 412 . ”; Col. 8 Ln. 56-59 – “The computing device 600 can generate and display a GUI that includes an alert indicating whether the particular areas are suitable for further exploration or should be avoided during a drilling operation. ” ]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to generate the control signal/alert of Rueger following the subsurface fault location detection of Zheng and Zhang in order to signify locations to drill into the Earth. Regarding Claim 18, the combination of Zhang and Zheng would disclose the one or more non-transitory computer readable media of claim 17. The combination does not disclose the operations further comprising: based on the prediction value and the uncertainty value, generating a control signal configured for causing drilling of a well in the subsurface corresponding to the location in the seismic image. Rueger, however, discloses the operations further comprising: based on the prediction value and the uncertainty value, generating a control signal configured for causing drilling of a well in the subsurface corresponding to the location in the seismic image [ Col. 8, Ln. 29-39 - “For example, the computing device 600 may execute the action module 618 to control a drilling operation within a subterranean formation . In such an example, a drilling direction may be determined based on a predicted location of a fault in the subterranean formation prior to drilling into the fault. The control of the drilling operation may also generally include making a determination about whether to drill a well based on economic risk implied by the volumetric uncertainty provided by the fault uncertainty confidence indexes 412 . ”; Col. 8 Ln. 56-59 – “The computing device 600 can generate and display a GUI that includes an alert indicating whether the particular areas are suitable for further exploration or should be avoided during a drilling operation. ” ]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to generate the control signal/alert of Rueger following the subsurface fault location detection of Zheng and Zhang in order to signify locations to drill into the Earth. Claims 4, 14, and 19 are rejected under Zhang et. al in view of Zheng et. al, in further view of Evans et. al. (US 11113632 B1). Regarding Claim 4, Zhang and Zheng disclose the method of claim 1, wherein the data processing model is trained using one or more seismic images labeled with fault or fracture labels [ Zhang, Paragraph [0054] – “This process is repeated for additional images 88 (e.g., additional voxels of the seismic image being processed) until the seismic image of interest is processed to reveal the faults present therein.” – see Fig. [6]; Paragraph [0062] – “In present embodiments, the training data 110 , training data 112 , and training data 114 is 3D synthetic training data; however, actual recorded data, for example, from previous expeditions could be used in place of or in conjunction with the synthetic data . Benefits from the use of synthetic data for training include no human labeling required , reduction/elimination of manually labeled fault dips and azimuths in 3D field data, unlimited possibilities for the number of training data and labels, ease in populating all possible fault dips and azimuths, known ground truth labels, avoidance of existing manual selections that often following fault truncations inaccurately (rendering them inadequate for training).” – actual recorded data is labeled ], The combination does not disclose wherein the training comprises maximizing an evidence lower bound value based on a stochastic gradient ascent. However, Evans discloses wherein the training comprises maximizing an evidence lower bound value based on a stochastic gradient ascent [ Col. 7, Ln. 7-31 – “ In variational inference , a form of the posterior defined by a set of free variational parameters is assumed, which is estimated by minimizing the Kullback-Leibler divergence from this assumed variational posterior to the exact posterior. Minimizing the Kullback-Leibler divergence can be equivalent to maximization of the evidence lower bound (ELBO) and therefore the objective in 412 refers to the ELBO in the case of variational inference . Assuming that the likelihood is independent between the n.sub.data observations, the ELBO can be determined as a sum over the n.sub.data train ing observation s; enabling the use of mini-batch sampling to be used for stochastic gradient computation, and large datasets to be considered. In the mini-batch sampling case, only a subset of the train ing dataset would be used in 404 at each iteration. Additional stochasticity may also be used to estimate expectations present in the ELBO (in situations where the terms cannot be computed in closed-form) through the use of a reparameterization approach, REINFORCE, or both. These computational strategies enable the use of gradient-based optimization techniques (such as stochastic gradient ascent) to be employed to maximize the ELBO with respect to parameters (such as variational parameters), and thus perform variational inference.” – note use of training observations ]. It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to train the data processing model using the seismic images of Zhang and Zheng using the variational techniques of Evans in order to reduce computational costs. Regarding Claim 14, Zhang and Zheng disclose the system of claim 11, wherein the data processing model is trained using one or more seismic images labeled with fault or fracture labels [ Zhang, Paragraph [0054] – “This process is repeated for additional images 88 (e.g., additional voxels of the seismic image being processed) until the seismic image of interest is processed to reveal the faults present therein.” – see Fig. [6]; Paragraph [0062] – “In present embodiments, the training data 110 , training data 112 , and training data 114 is 3D synthetic training data; however, actual recorded data, for example, from previous expeditions could be used in place of or in conjunction with the synthetic data . Benefits from the use of synthetic data for training include no human labeling required , reduction/elimination of manually labeled fault dips and azimuths in 3D field data , unlimited possibilities for the number of training data and labels, ease in populating all possible fault dips and azimuths, known ground truth labels, avoidance of existing manual selections that often following fault truncations inaccurately (rendering them inadequate for training).” – actual recorded data is labeled ], The combination does not disclose wherein the training comprises maximizing an evidence lower bound value based on a stochastic gradient ascent. However, Evans discloses wherein the training comprises maximizing an evidence lower bound value based on a stochastic gradient ascent [ Col. 7, Ln. 7-31 – “ In variational inference , a form of the posterior defined by a set of free variational parameters is assumed, which is estimated by minimizing the Kullback-Leibler divergence from this assumed variational posterior to the exact posterior. Minimizing the Kullback-Leibler divergence can be equivalent to maximization of the evidence lower bound (ELBO) and therefore the objective in 412 refers to the ELBO in the case of variational inference . Assuming that the likelihood is independent between the n.sub.data observations, the ELBO can be determined as a sum over the n.sub.data train ing observations ; enabling the use of mini-batch sampling to be used for stochastic gradient computation , and large datasets to be considered. In the mini-batch sampling case, only a subset of the train ing dataset would be used in 404 at each iteration. Additional stochasticity may also be used to estimate expectations present in the ELBO (in situations where the terms cannot be computed in closed-form) through the use of a reparameterization approach, REINFORCE, or both. These computational strategies enable the use of gradient-based optimization techniques (such as stochastic gradient ascent) to be employed to maximize the ELBO with respect to parameters (such as variational parameters), and thus perform variational inference.” – note use of training observations ]. It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to train the data processing model using the seismic images of Zhang and Zheng using the variational techniques of Evans in order to reduce computational costs. Regarding Claim 19, Zhang and Zheng disclose the one or more non-transitory computer readable media of claim 17, wherein the data processing model is trained using one or more seismic images labeled with fault or fracture labels [ Zhang, Paragraph [0054] – “This process is repeated for additional images 88 (e.g., additional voxels of the seismic image being processed) until the seismic image of interest is processed to reveal the faults present therein.” – see Fig. [6]; Paragraph [0062] – “In present embodiments, the training data 110 , training data 112 , and training data 114 is 3D synthetic training data; however, actual recorded data, for example, from previous expeditions could be used in place of or in conjunction with the synthetic data . Benefits from the use of synthetic data for training include no human labeling required , reduction/elimination of manually labeled fault dips and azimuths in 3D field data, unlimited possibilities for the number of training data and labels, ease in populating all possible fault dips and azimuths, known ground truth labels, avoidance of existing manual selections that often following fault truncations inaccurately (rendering them inadequate for training).” – actual recorded data is labeled ], The combination does not disclose wherein the training comprises maximizing an evidence lower bound value based on a stochastic gradient ascent. However, Evans discloses wherein the training comprises maximizing an evidence lower bound value based on a stochastic gradient ascent [ Col. 7, Ln. 7-31 – “ In variational inference , a form of the posterior defined by a set of free variational parameters is assumed, which is estimated by minimizing the Kullback-Leibler divergence from this assumed variational posterior to the exact posterior. Minimizing the Kullback-Leibler divergence can be equivalent to maximization of the evidence lower bound (ELBO) and therefore the objective in 412 refers to the ELBO in the case of variational inference . Assuming that the likelihood is independent between the n.sub.data observations, the ELBO can be determined as a sum over the n.sub.data train ing observations; enabling the use of mini-batch sampling to be used for stochastic gradient computation , and large datasets to be considered. In the mini-batch sampling case, only a subset of the train ing dataset would be used in 404 at each iteration. Additional stochasticity may also be used to estimate expectations present in the ELBO (in situations where the terms cannot be computed in closed-form) through the use of a reparameterization approach, REINFORCE, or both. These computational strategies enable the use of gradient-based optimization techniques (such as stochastic gradient ascent) to be employed to maximize the ELBO with respect to parameters (such as variational parameters), and thus perform variational inference.” – note use of training observations ]. It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to train the data processing model using the seismic images of Zhang and Zheng using the variational techniques of Evans in order to reduce computational costs. Claims 8 and 9 are rejected under Zhang et. al in view of Zheng et. al in further view of Colombo et. al. (US 20230288592 A1). Regarding Claim 8, Zhang and Zheng would disclose method of claim 1. The combination does not disclose wherein a machine learning parameter of the one or more machine learning parameters comprises a neural network bias value. However, Colombo discloses wherein a machine learning parameter of the one or more machine learning parameters comprises a neural network bias value [ Paragraph [0040] – “ For a neural network ( 200 ) to complete a “task” of predicting output data from observed input data, the neural network ( 200 ) must first be trained…Backpropagation is defined as using a gradient descent algorithm to update the weights and bias terms within a neural network ( 200 ). ” ]. It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to use the bias value of Colombo as one of the machine learning parameters of Zhang and Zheng in order to facilitate convergence on the posterior distribution. Regarding Claim 9, Zhang and Zheng would disclose method of claim 1. The combination does not disclose wherein a machine learning parameter of the one or more machine learning parameters comprises a neural network weight value. Colombo, however, discloses wherein a machine learning parameter of the one or more machine learning parameters comprises a neural network weight value [ Paragraph [0040] – “ For a neural network ( 200 ) to complete a “task” of predicting output data from observed input data, the neural network ( 200 ) must first be trained…Backpropagation is defined as using a gradient descent algorithm to update the weights and bias terms within a neural network ( 200 ). ” ]. It would have been obvious to one of ordinary skill in the art, before the effective filing date of the claimed invention, to use the bias value of Colombo as one of the machine learning parameters of Zhang and Zheng in order to facilitate convergence on the posterior distribution. Pertinent Prior Art 07-96 AIA The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Iqbal, N., SYSTEM AND METHOD FOR LIGHTWEIGHT IN-FIELD COMPRESSION OF SEISMIC DATA , US 20250030435 A1, 2025. Zhang, J., Robust Stochastic Seismic Inversion With New Error Term Specification , US 20240125958 A1, 2024. Chen, J. , SYSTEM AND METHOD FOR SEISMIC DEPTH UNCERTAINTY ANALYSIS , US 20230288593 A1, 2023. Li, D., METHOD AND SYSTEM FOR AUGMENTED INVERSION AND UNCERTAINTY QUANTIFICATION FOR CHARACTERIZING GEOPHYSICAL BODIES, US 20230032044 A1, 2023. Jose Bernardo, Bayesian Statistics, Encyclopedia of Life Support Systems, 2003, Probability and Statistics (R. Viertl, ed) of the Encyclopedia of Life Support Systems (EOLSS) . Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to JANELLE A HOLMES whose telephone number is (571)272-4336. The examiner can normally be reached Monday - Friday 8:00 m - 5 pm. 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, Arleen M Vazquez can be reached at (571) 272-2619. 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. /KYLE R QUIGLEY/Primary Examiner, Art Unit 2857 /J.A.H./ Examiner, Art Unit 2857 Application/Control Number: 18/404,353 Page 2 Art Unit: 2857 Application/Control Number: 18/404,353 Page 3 Art Unit: 2857 Application/Control Number: 18/404,353 Page 4 Art Unit: 2857 Application/Control Number: 18/404,353 Page 5 Art Unit: 2857 Application/Control Number: 18/404,353 Page 6 Art Unit: 2857 Application/Control Number: 18/404,353 Page 7 Art Unit: 2857 Application/Control Number: 18/404,353 Page 8 Art Unit: 2857 Application/Control Number: 18/404,353 Page 9 Art Unit: 2857 Application/Control Number: 18/404,353 Page 10 Art Unit: 2857 Application/Control Number: 18/404,353 Page 11 Art Unit: 2857 Application/Control Number: 18/404,353 Page 12 Art Unit: 2857 Application/Control Number: 18/404,353 Page 13 Art Unit: 2857 Application/Control Number: 18/404,353 Page 14 Art Unit: 2857 Application/Control Number: 18/404,353 Page 15 Art Unit: 2857 Application/Control Number: 18/404,353 Page 16 Art Unit: 2857 Application/Control Number: 18/404,353 Page 17 Art Unit: 2857 Application/Control Number: 18/404,353 Page 18 Art Unit: 2857 Application/Control Number: 18/404,353 Page 19 Art Unit: 2857 Application/Control Number: 18/404,353 Page 20 Art Unit: 2857 Application/Control Number: 18/404,353 Page 21 Art Unit: 2857 Application/Control Number: 18/404,353 Page 22 Art Unit: 2857 Application/Control Number: 18/404,353 Page 23 Art Unit: 2857 Application/Control Number: 18/404,353 Page 24 Art Unit: 2857 Application/Control Number: 18/404,353 Page 25 Art Unit: 2857 Application/Control Number: 18/404,353 Page 26 Art Unit: 2857 Application/Control Number: 18/404,353 Page 27 Art Unit: 2857 Application/Control Number: 18/404,353 Page 28 Art Unit: 2857 Application/Control Number: 18/404,353 Page 29 Art Unit: 2857 Application/Control Number: 18/404,353 Page 30 Art Unit: 2857
Prosecution Insights