DETAILED ACTION
1. This communication is in response to the request for continued examination filed on March 10, 2026 and the corresponding amendments filed on February 3, 2026 for Application No. 17/664,961 in which Claims 21-22, 25-26, and 29-32 are presented for examination.
Notice of Pre-AIA or AIA Status
2. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Continued Examination Under 37 CFR 1.114
3. A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 03/10/2026 has been entered.
Response to Arguments
4. The amendments filed on February 3, 2026 have been considered. Claims 21, 25, and 29 have been amended. Claims 23-24, 27-28, and 33-34 have been cancelled. Thus, Claims 21-22, 25-26, and 29-32 are pending and presented for examination.
5. Applicant’s arguments filed February 3, 2026 with respect to the 35 U.S.C. 112(b) rejection have been fully considered and are persuasive. Thus, the 35 U.S.C. 112(b) rejection has been withdrawn.
6. Applicant's arguments filed February 3, 2026 with respect to the 35 U.S.C. 101 rejection have been fully considered but they are not persuasive.
Applicant’s Arguments on Pg. 10 of Arguments/Remarks state:
“Though the Examiner suggested that Applicant not delete the deleted limitations during the examiner interview, Applicant notes that the Examiner contends that the recited residual functions are mathematical concepts. See final Office action at 8. Accordingly, it is unclear to Applicant why maintaining the deleted limitations would be beneficial in overcoming the § 101 rejection. In contrast to what the Examiner contends, the lack of recitation of any alleged judicial exceptions causes amended independent claims 21, 25, and 29 to not be directed to any alleged judicial exceptions under Step 2A Prong One. Such a conclusion renders amended independent claims 21, 25, and 29 patent eligible.”
Examiner respectfully disagrees. As stated during the Interview conducted on 12/22/2025 (and according Interview Summary dated 12/29/2025), Examiner recommended to maintain the “deleted limitations” to avoid potentially deleting/removing any previously indicated allowable subject matter, at the time in which these amendments were discussed. Upon performing an updated search on the amended limitations, Examiner maintains that Claims 21-22, 25-26, and 29-32 still indicate allowable matter, although they are still rejected by 35 U.S.C. 101 – abstract idea.
Further, as also stated in the Interview, Examiner reiterates to Applicant that generically stating “training, using backpropagation and a first/second residual function, the first/second neural network […]” does not provide an inventive concept – indeed, these steps would be considered merely a combination of “apply it” training and mere recitation of mathematical processes/concepts without significantly more. Examiner suggested during the Interview to further amend any “training” limitations to better illustrate how the “training” reflects an improvement on the neural networks/system as a whole. Otherwise, as currently drafted, Applicant’s claims merely recite obtaining multiple datasets and simply training a first/second generic (e.g., black box) neural networks by merely using these obtained datasets without significantly more. Thus, Examiner recommended maintaining the “deleted limitations” but suggested further claim amendments which would have clarified the technical processes used during training, in order to integrate the judicial exception into a practical application at Step 2A Prong 2 and include additional elements that are sufficient to amount to significantly more than the judicial exception at Step 2B.
Applicant’s Arguments on Pgs. 11-12 of Arguments/Remarks state:
“In the USPTO's analysis of the claim, the USPTO, with respect to Step 1 of the SME test, states that "Yes. The claim recites a series of steps and, therefore, is a process." Subject Matter Eligibility Examples: Abstract Ideas 37 - 42 at 9. The USPTO, with respect to Step 2A Prong One of the SME test, states that "No. The claim does not recite any of the judicial exceptions enumerated in the 2019 PEG. For instance, the claim does not recite any mathematical relationships, formulas, or calculations. While some of the limitations may be based on mathematical concepts, the mathematical concepts are not recited in the claims. Further, the claim does not recite a mental process because the steps are not practically performed in the human mind . .. Thus, the claim is eligible because it does not recite a judicial exception." Id. (emphasis added).
The USPTO-issued August 4, 2025, memorandum on reminders on evaluating subject matter eligibility of claims under 35 U.S.C. § 101 goes on to state, in reference to the claim of USPTO- issued Example 39, that "[t]he claim limitation "training the neural network in a first stage using the first training set" of example 39 does not recite a judicial exception. Even though "training the neural network" involves a broad array of techniques and/or activities that may involve or rely upon mathematical concepts, the limitation does not set forth or describe any mathematical relationships, calculations, formulas, or equations using words or mathematical symbols. Contrast this with the limitation "training, by the computer, the ANN based on the input data and a selected training algorithm to generate a trained ANN, wherein the selected training algorithm includes a backpropagation algorithm and a gradient descent algorithm" of claim 2 of [USPTO-issued] example 47. This limitation requires specific mathematical calculations by referring to the mathematical calculations by name, i.e., a backpropagation algorithm and a gradient descent algorithm, and therefore recites a judicial exception, namely an abstract idea." August 4, 2025, memorandum at 3.
However, the Examiner contends that "the steps preceding the 'training' limitations of Example 39, clearly show an improvement on collected/obtained data and also cannot be practically performed by mental process" and is why the claim of Example 39 is patent eligible. Applicant- Initiated Interview Summary dated December 29, 2025, at 2. In view of the above, the Examiner's contention for why USPTO-issued Example 39 is patent eligible appears ill-founded and unsupported by the USPTO in both the Subject Matter Eligibility Examples: Abstract Ideas 37 - 42 and August 4, 2025, memorandum.”
Examiner respectfully disagrees. First, it is unclear what exactly Applicant is referencing regarding “Examiner’s contention for why USPTO-issued example 39 is patent eligible appears ill-founded and unsupported by the USPTO […]” – Examiner stated that Example 39 is eligible for the same reasons as cited by Applicant on Arguments/Remarks Pg. 11. Example 39 does not merely recite “collecting” or obtaining data and inputting this collected/obtained data into a neural network for generic “training”. Instead, as also explained during the Interview, Example 39 recites the limitations “applying one or more transformations to each digital facial image including mirroring, rotating, smoothing, or contrast reduction to create a modified set of digital facial images” and “creating a first training set comprising the collected set of digital facial images, the modified set of digital facial images, and a set of digital non-facial images” which clearly cannot be performed by mental process or mathematical process alone at Step 2A Prong 1, as a human user cannot feasibly apply transformations to digital facial images and create a training set comprising digital facial images by mental process. In comparison to the instant claim limitations, Independent Claim 21 recites “producing predicted second performance data […] in response to the subset of the second geoscience data” and “determining a residual between the second performance data and the predicted second performance data” – these steps may clearly be performed by mental process at Step 2A Prong 1. For example, a human is capable of observing/analyzing a subset of second geoscience data and using judgement/evaluation to produce predicted second performance data (an estimate or prediction of a well’s initial production rate/production decline rate/total cumulative hydrocarbon production/etc.) based on said analysis of the second geoscience data (which may include geological/petrophysical/well completion data). Further, a human user is also capable of observing/analyzing the obtained second performance data and the predicted second performance data and using judgement/evaluation to determine a residual or a difference (See Applicant’s specification Par. [0050] for support on the residual comprising a difference) between the performance data/metrics based on said analysis. Thus, the instant claims draw no parallels to Example 39 and therefore, the instant claims are ineligible.
Furthermore, it is not entirely clear what Applicant is referencing regarding the citation of Claim 2 of Example 47, including the limitation “training, by the computer, the ANN based on the input data and a selected training algorithm to generate a trained ANN, wherein the selected training algorithm includes a backpropagation algorithm and a gradient descent algorithm” as Claim 2 of Example 47 is ineligible. In fact, Examiner contends that the instant claims are ineligible for substantially the same reasons as presented by Claim 2 of SME Example 47.
Applicant’s Arguments on Pgs. 12-14 of Arguments/Remarks state:
“Applicant also submits that amended independent claims 21 and 25 are similar to the claims of Ex Parte Desjardins, Appeal No. 2024-000567. The USPTO-issued December 5, 2025, memorandum on advance notice of change to the MPEP in light of Ex Parte Desjardins notes that the claims of Ex Parte Desjardins are patent eligible because "the specification identified improvements as to how the machine learning model itself operates." December 5, 2025, memorandum at 2. Note that the claims of Ex Parte Desjardins are drawn to a computer-implemented method of training machine learning models. Similarly, the neural networks recited in independent claims 21, 25, and 29 themselves operate to improve on transfer learning such that additional data types may be used for prediction purposes in contrast to traditional transfer learning. Traditional transfer learning typically uses the assigned values of weights of one trained neural network that is trained to perform a task as a starting point for a second neural network that is to be trained to perform another task. In doing so, only a subset of weights of the second neural network may need to be retrained to thus reduce the computational expense of training the second neural network. See Applicant's specification, para. 49. Such a "knowledge transfer" typically requires that the data input into both the neural networks and the predictions produced from the neural networks be of the same data types. In contrast to traditional transfer learning, the recited methods of amended independent claims 21, 25, and 29 transfer knowledge even when additional data types are available to be used for predictions by way of the second neural network producing a new residual in response to new geoscience data (that includes the additional data types recited as the "second set of data types") and applying the new residual to predicted new performance data produced from the first neural network such that the updated predicted new performance data accounts for the additional data types of the new geoscience data. To do so, the second neural network is trained using, in part, a residual that is determined during the training of the first neural network. A residual thus transfers knowledge rather than transferring knowledge through weights of a trained neural network as is done in traditional transfer learning.
In view of the above, amended independent claims 21, 25, and 29 clearly reflect the improvement on transfer learning by virtue of the first neural network being trained using a subset of the second geoscience data (that comprises the first set of data types), the second neural network being trained using all the second geoscience data, the trained first neural network producing predicted new performance data in response to the subset of the new geoscience data, the trained second neural network producing a new residual in response to all the new geoscience data, and applying the new residual to the predicted new performance data to determine updated predicted new performance data.”
Examiner respectfully disagrees. First, Examiner notes that the instant claims no longer explicitly recite the use of any transfer learning processes, as this language has now been removed from the claims in light of the most recently filed amendments. Furthermore, in contrast to Ex Parte Desjardins in which “the specification identified improvements as to how the machine learning model itself operates”, the supposed improvements provided to such transfer learning processes are not at all cited by Applicant’s specification. The only mention of the term “transfer learning” appears in Applicant’s specification Par. [0049] and states “Re-training may use a transfer learning approach. Transfer learning is defined as re-training a subset of weights within the arrays of weights and a subset of the bias terms within a trained neural network (300). Similar to training, re-training may be performed iteratively using training data and backpropagation.” – however, again, this does not recite any technical improvement to the transfer learning process. Moreover, with regards to the “residual” that is used to train the networks, this amounts to no more than a mere difference between the sets of performance data and predicted performance data (as supported by Applicant’s specification Par. [0044]) and/or the use of a residual function (which may also comprise an objective function, cost function, loss function, or error function per Applicant’s specification Par. [0039]). Furthermore, the neural networks are merely trained using the residual without significantly more – this does not provide an inventive concept. Applicant is encouraged to provide more technical detail in the claim language, in order to highlight the technical improvement provided by “training” limitations and the use of the residual value, such that the “training” is not deemed to be merely “applied” or adding the words “apply it” to the judicial exception without significantly more.
Thus, for the reasons stated above, the 35 U.S.C. 101 rejection is maintained.
7. Applicant’s arguments filed February 3, 2026 with respect to the 35 U.S.C. 103 rejection have been fully considered and are persuasive. Thus, the 35 U.S.C. 103 rejection has been withdrawn. Examiner’s Note: No prior art rejection is made for Claims 21-22, 25-26, and 29-32. However, these claims are still rejected under 35 U.S.C. 101 – abstract idea.
Claim Rejections - 35 USC § 101
8. 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.
9. Claims 21-22, 25-26, and 29-32 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding Claim 21:
Step 1: Claim 21 is a method type claim. Therefore, Claims 21-22 are directed to either a process, machine, manufacture, or composition of matter.
2A Prong 1: If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation by mathematical calculation but for the recitation of generic computer components, then it falls within the “Mathematical Concepts” grouping of abstract ideas.
producing predicted second performance data […] in response to the subset of the second geoscience data (mental process – other than reciting “first neural network”, producing predicted second performance data may be performed manually by a user observing/analyzing the subset of second geoscience data (geological/petrophysical/well completion data) and accordingly using judgement/evaluation to cast a prediction regarding second performance data (production rate/production decline rate/total cumulative hydrocarbon production) based on the analysis of the subset of second geoscience data)
determining a residual between the second performance data and the predicted second performance data (mental process/mathematical process – determining a residual between the second performance data may be performed manually by a user observing/analyzing the second performance data and predicted second performance data and accordingly using judgement/evaluation to compare and/or determine a difference between the sets of data (See Applicant’s specification Par. [0050] which supports this interpretation). Further, this may also be completed by mathematical process utilizing a residual function (or cost/loss/objective function – See Applicant’s specification Par. [0039]) to determine such a residual/difference)
2A Prong 2: This judicial exception is not integrated into a practical application.
Additional elements:
method of training a first neural network and a second neural network (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of training a machine learning model with previously determined/obtained data without significantly more)
obtaining first geoscience data and first performance data for a first set of wells, wherein the first geoscience data comprise a first set of data types (Adding insignificant extra-solution activity to the judicial exception – see MPEP 2106.05(g))
obtaining second geoscience data and second performance data for a second set of wells (Adding insignificant extra-solution activity to the judicial exception – see MPEP 2106.05(g))
wherein the second geoscience data comprise a second set of data types, wherein a subset of the second geoscience data comprises the first set of data types (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the second geoscience data comprises a second set of data types wherein a subset of the second data comprises a first set of data types does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
wherein each of the first performance data and the second performance data comprises at least one of an initial production rate, a production decline rate, a bottom- hole pressure decline, a total cumulative hydrocarbon production over a time window, or an estimate ultimate recovery (EUR) of each of the first set of wells and the second set of wells(Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying the first/second performance data may include a generic set of metrics does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
training the first neural network using the first geoscience data, the subset of the second geoscience data, the first performance data, and the second performance data (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of training a machine learning model with previously determined/obtained data without significantly more)
inputting the subset of the second geoscience data into the first neural network (Adding insignificant extra-solution activity to the judicial exception – see MPEP 2106.05(g))
wherein the first neural network is trained to produce predicted new performance data in response to a subset of new geoscience data input into the first neural network (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of training a machine learning model with previously determined/obtained data without significantly more)
wherein the subset of the new geoscience data comprises the first set of data types (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the subset of new geoscience data comprises the first set of data types does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
training the second neural network using the second geoscience data and the residual, wherein the second neural network is trained to produce a new residual in response to the new geoscience data input into the second neural network (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of training a machine learning model with previously determined/obtained data without significantly more)
wherein the new geoscience data comprise the second set of data types (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the new geoscience data comprises the second set of data types does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
method of training a first neural network and a second neural network (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of training a machine learning model with previously determined/obtained data without significantly more. This cannot provide an inventive concept)
obtaining first geoscience data and first performance data for a first set of wells, wherein the first geoscience data comprise a first set of data types (MPEP 2106.05(d)(II) indicates that merely “Receiving or transmitting data over a network” is a well-understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim). Thereby, a conclusion that the claimed limitation is well-understood, routine, conventional activity is supported under Berkheimer)
obtaining second geoscience data and second performance data for a second set of wells (MPEP 2106.05(d)(II) indicates that merely “Receiving or transmitting data over a network” is a well-understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim). Thereby, a conclusion that the claimed limitation is well-understood, routine, conventional activity is supported under Berkheimer)
wherein the second geoscience data comprise a second set of data types, wherein a subset of the second geoscience data comprises the first set of data types (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the second geoscience data comprises a second set of data types wherein a subset of the second data comprises a first set of data types does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
wherein each of the first performance data and the second performance data comprises at least one of an initial production rate, a production decline rate, a bottom- hole pressure decline, a total cumulative hydrocarbon production over a time window, or an estimate ultimate recovery (EUR) of each of the first set of wells and the second set of wells(Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying the first/second performance data may include a generic set of metrics does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
training the first neural network using the first geoscience data, the subset of the second geoscience data, the first performance data, and the second performance data (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of training a machine learning model with previously determined/obtained data without significantly more. This cannot provide an inventive concept)
inputting the subset of the second geoscience data into the first neural network (MPEP 2106.05(d)(II) indicates that merely “Receiving or transmitting data over a network” is a well-understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim). Thereby, a conclusion that the claimed limitation is well-understood, routine, conventional activity is supported under Berkheimer)
wherein the first neural network is trained to produce predicted new performance data in response to a subset of new geoscience data input into the first neural network (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of training a machine learning model with previously determined/obtained data without significantly more. This cannot provide an inventive concept)
wherein the subset of the new geoscience data comprises the first set of data types (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the subset of new geoscience data comprises the first set of data types does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
training the second neural network using the second geoscience data and the residual, wherein the second neural network is trained to produce a new residual in response to the new geoscience data input into the second neural network (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of training a machine learning model with previously determined/obtained data without significantly more. This cannot provide an inventive concept)
wherein the new geoscience data comprise the second set of data types (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the new geoscience data comprises the second set of data types does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
For the reasons above, Claim 21 is rejected as being directed to an abstract idea without significantly more. This rejection applies equally to dependent claim 22. The additional limitations of the dependent claims are addressed below.
Regarding Claim 22:
Step 2A Prong 1:
See the rejection of Claim 21 above, which Claim 22 depends on.
Step 2A Prong 2 & Step 2B:
wherein each of the first geoscience data and the second geoscience data comprises well log data (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the first/second geoscience data comprises well log data does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, these additional elements do not integrate the abstract idea into practical application because they do not impose any meaningful limits on practicing the abstract idea, as discussed above in the rejection of claim 21.
Independent Claim 25 recites substantially the same limitations as Claim 21, in the form of a non-transitory computer-readable memory, including generic computer components. The claim is also directed to performing mental processes/mathematical calculations without significantly more, therefore it is rejected under the same rationale.
For the reasons above, Claim 25 is rejected as being directed to an abstract idea without significantly more. This rejection applies equally to dependent claim 26. The additional limitations of the dependent claims are addressed below.
Claim 26 recites substantially the same limitations as Claim 22, in the form of a non-transitory computer-readable memory, including generic computer components. The claim is also directed to performing mental processes/mathematical calculations without significantly more, therefore it is rejected under the same rationale.
Regarding Claim 29:
Step 1: Claim 29 is a method type claim. Therefore, Claims 29-32 are directed to either a process, machine, manufacture, or composition of matter.
2A Prong 1: If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation by mathematical calculation but for the recitation of generic computer components, then it falls within the “Mathematical Concepts” grouping of abstract ideas.
producing predicted new performance data for the new well […] in response to the subset of the new geoscience data (mental process – other than reciting “trained first neural network”, producing predicted new performance data for the new well may be performed manually by a user observing/analyzing the subset of new geoscience data (geological/petrophysical/well completion data) and accordingly using judgement/evaluation to cast a prediction regarding new performance data (production rate/production decline rate/total cumulative hydrocarbon production) for the new well based on the analysis of the new geoscience data)
producing a new residual […] in response to the new geoscience data (mental process – other than reciting “trained second neural network”, producing a new residual may be performed manually by a user observing/analyzing the new geoscience data and accordingly using judgement/evaluation to produce a new residual/difference (See Applicant’s specification Par. [0050] which supports this interpretation) in response to said analysis of the new geoscience data. Alternatively, producing a new residual may be performed by mathematical process using a residual function (or loss/objective/cost function - See Applicant’s specification Par. [0039]))
determining updated predicted new performance data for the new well by applying the new residual to the predicted new performance data (mental process – determining updated predicted new performance data may be performed manually by a user observing/analyzing the residual/difference and predicted new performance data and accordingly using judgement/evaluation to apply the new residual/difference to the predicted new performance data (by comparing both sets of data with the aid of pen and paper), hence determining updated predicted new performance data for the new well)
2A Prong 2: This judicial exception is not integrated into a practical application.
Additional elements:
obtaining new geoscience data for a new well (Adding insignificant extra-solution activity to the judicial exception – see MPEP 2106.05(g))
inputting a subset of the new geoscience data into a trained first neural network (Adding insignificant extra-solution activity to the judicial exception – see MPEP 2106.05(g))
wherein the subset of the new geoscience data comprises a first set of data types and wherein the new geoscience data comprise a second set of data types (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the subset of new geoscience data comprises a first/second set of data types does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
[…] trained first neural network […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of applying a trained machine learning model without significantly more)
wherein the predicted performance data comprises at least one of an initial production rate, a production decline rate, a bottom-hole pressure decline, a total cumulative hydrocarbon production over a time window, or an estimate ultimate recovery (EUR) of the new well (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the predicted performance data may include a generic set of metrics does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
inputting the new geoscience data into a trained second neural network (Adding insignificant extra-solution activity to the judicial exception – see MPEP 2106.05(g))
[…] trained second neural network […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of applying a trained machine learning model without significantly more)
2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Additional elements:
obtaining new geoscience data for a new well (MPEP 2106.05(d)(II) indicates that merely “Receiving or transmitting data over a network” is a well-understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim). Thereby, a conclusion that the claimed limitation is well-understood, routine, conventional activity is supported under Berkheimer)
inputting a subset of the new geoscience data into a trained first neural network (MPEP 2106.05(d)(II) indicates that merely “Receiving or transmitting data over a network” is a well-understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim). Thereby, a conclusion that the claimed limitation is well-understood, routine, conventional activity is supported under Berkheimer)
wherein the subset of the new geoscience data comprises a first set of data types and wherein the new geoscience data comprise a second set of data types (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the subset of new geoscience data comprises a first/second set of data types does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
[…] trained first neural network […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of applying a trained machine learning model without significantly more)
wherein the predicted performance data comprises at least one of an initial production rate, a production decline rate, a bottom-hole pressure decline, a total cumulative hydrocarbon production over a time window, or an estimate ultimate recovery (EUR) of the new well (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the predicted performance data may include a generic set of metrics does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
inputting the new geoscience data into a trained second neural network (MPEP 2106.05(d)(II) indicates that merely “Receiving or transmitting data over a network” is a well-understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim). Thereby, a conclusion that the claimed limitation is well-understood, routine, conventional activity is supported under Berkheimer)
[…] trained second neural network […] (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: high level recitation of applying a trained machine learning model without significantly more)
For the reasons above, Claim 29 is rejected as being directed to an abstract idea without significantly more. This rejection applies equally to dependent claims 30-32. The additional limitations of the dependent claims are addressed below.
Regarding Claim 30:
Step 2A Prong 1: See the rejection of Claim 29 above, which Claim 30 depends on.
completing, using well completion operations, the new well based on the updated predicted new performance data (mental process – completing the new well may be performed manually by a user observing/analyzing the updated predicted new performance data and accordingly using judgement/evaluation to determine a hydrocarbon field management plan for completing the new well based on the updated predicted new performance data (See claim language of Claim 31 for recitation of where completing the new well comprises determining a hydrocarbon field management plan))
Step 2A Prong 2 & Step 2B:
Accordingly, under Step 2A Prong 2 and Step 2B, these additional elements do not integrate the abstract idea into practical application because they do not impose any meaningful limits on practicing the abstract idea, as discussed above in the rejection of claim 29.
Regarding Claim 31:
Step 2A Prong 1: See the rejection of Claim 30 above, which Claim 31 depends on.
wherein completing the new well comprises determining a hydrocarbon field management plan based on the updated predicted new performance data (mental process – determining a hydrocarbon field management plan may be performed manually by a user observing/analyzing the updated predicted new performance data and accordingly using judgement/evaluation to determine a hydrocarbon field management plan (See Applicant’s specification Par. [0056]) based on said analysis)
Step 2A Prong 2 & Step 2B:
Accordingly, under Step 2A Prong 2 and Step 2B, these additional elements do not integrate the abstract idea into practical application because they do not impose any meaningful limits on practicing the abstract idea, as discussed above in the rejection of claim 29.
Regarding Claim 32:
Step 2A Prong 1:
See the rejection of Claim 29 above, which Claim 32 depends on.
Step 2A Prong 2 & Step 2B:
wherein the new geoscience data comprises new well log data (Field of Use – limitations that amount to merely indicating a field of use or technological environment in which to apply a judicial exception does not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application; in this case specifying that the new geoscience data comprises new well log data does not integrate the exception into a practical application nor amount to significantly more – See MPEP 2106.05(h))
Accordingly, under Step 2A Prong 2 and Step 2B, these additional elements do not integrate the abstract idea into practical application because they do not impose any meaningful limits on practicing the abstract idea, as discussed above in the rejection of claim 29.
Allowable Subject Matter
10. No prior art rejection is made for Claims 21-22, 25-26, and 29-32. However, these claims are still rejected under 35 U.S.C. 101 – abstract idea.
11. Examiner has disclosed Bansal et al. (US PG-PUB 20230193754) and Srinivasan et al. (“A machine learning framework for rapid forecasting and history matching in unconventional reservoirs”), which are the closest prior art as compared to the instant application. Bansal discloses machine learning assisted parameter matching and production forecasting for new wells – more specifically, Bansal discloses a first machine learning model trained to predict well logs for existing production wells, a first well model to estimate production of the existing production wells based on predicted well logs, and a second machine learning model to predict parameters of a second well model for a new production well, based on the tuned parameters of the first well model, with the new well’s production being forecasted using the second machine learning model. Srinivasan discloses a machine learning framework for forecasting production in unconventional reservoirs, specifically utilizing a loss function to optimize model performance. However, Bansal and Srinivasan seemingly do not explicitly disclose the training of the first neural network and second neural network, as per the instant claim limitations, including the limitations “training the first neural network using the first geoscience data, the subset of the second geoscience data, the first performance data, and the second performance data, wherein training the first neural network comprises: inputting the subset of the second geoscience data into the first neural network, producing predicted second performance data from the first neural network in response to the subset of the second geoscience data, and determining a residual between the second performance data and the predicted second performance data, wherein the first neural network is trained to produce predicted new performance data in response to a subset of new geoscience data input into the first neural network, and wherein the subset of the new geoscience data comprises the first set of data types; and training the second neural network using the second geoscience data and the residual, wherein the second neural network is trained to produce a new residual in response to the new geoscience data input into the second neural network, and wherein the new geoscience data comprise the second set of data types.” of Independent Claim 21 (and Independent Claim 25 which recites substantially the same limitations), in combination with the remaining limitations of the Independent claims. Further, Bansal and Srinivasan seemingly do not explicitly disclose the method per the instant claim limitations, including “producing predicted new performance data for the new well from the trained first neural network in response to the subset of the new geoscience data, wherein the predicted new performance data comprises at least one of an initial production rate, a production decline rate, a bottom-hole pressure decline, a total cumulative hydrocarbon production over a time window, or an estimate ultimate recovery (EUR) of the new well; inputting the new geoscience data into a trained second neural network; producing a new residual from the trained second neural network in response to the new geoscience data; and determining updated predicted new performance data for the new well by applying the new residual to the predicted new performance data.” of Independent Claim 29, in combination with the remaining limitations of the Independent claims.
Conclusion
12. Any inquiry concerning this communication or earlier communications from the examiner should be directed to Devika S Maharaj whose telephone number is (571)272-0829. The examiner can normally be reached Monday - Thursday 8:30am - 5:30pm.
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, Alexey Shmatov can be reached at (571)270-3428. 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.
/DEVIKA S MAHARAJ/Examiner, Art Unit 2123