DETAILED ACTION
The Applicant’s response, received 20 July 2026, has been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of the Claims
Claims 1-20 are pending.
Claims 1-20 are rejected.
Priority
There are no domestic or foreign applications for which benefit is claimed.
Therefore, the effective filing date of the claimed invention is 15 November 2022.
Drawings
The objection to the drawings in the Office action mailed 06 May 2026 has been withdrawn in view of the amendment received 20 July 2026.
The replacement drawings received 20 July 2026 have been accepted.
Claim Objections
The objection to claim 11 in the Office action mailed 06 May 2026 has been withdrawn in view of the amendment received 20 July 2026.
Claim Rejections - 35 USC § 112
The rejection of claims 8, 16, and 17-20 under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, in the Office action mailed 06 May 2026 has been withdrawn in view of the amendment received 20 July 2026.
The Applicant’s amendment received 20 July 2026 has been fully considered, however after further consideration, new grounds of rejection are raised under 35 U.S.C. 112(d) or pre-AIA 35 U.S.C. 112, 4th paragraph, in view of the amendment.
The following is a quotation of 35 U.S.C. 112(d):
(d) REFERENCE IN DEPENDENT FORMS.—Subject to subsection (e), a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
The following is a quotation of pre-AIA 35 U.S.C. 112, fourth paragraph:
Subject to the following paragraph [i.e., the fifth paragraph of pre-AIA 35 U.S.C. 112], a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
Claims 2-5, 10-13, and 18 are rejected under 35 U.S.C. 112(d) or pre-AIA 35 U.S.C. 112, 4th paragraph, as being of improper dependent form for failing to further limit the subject matter of the claim upon which it depends, or for failing to include all the limitations of the claim upon which it depends.
Claim 2 recites “mapping comprises mapping the unit cell in the crystal structure to periodic boundary conditions.” Newly amended claim 1 recites this limitation, and therefore claim 2 is of improper dependent form for failing to further limit the subject matter of the claim upon which it depends.
Claim 3 recites “encoding the periodic boundary conditions as a Gaussian process.” Newly amended claim 1 recites this limitation, and therefore claim 3 is of improper dependent form for failing to further limit the subject matter of the claim upon which it depends.
Claim 4 recites “predicting the material property comprises interpolating the 3D embedding of the chemical elements proximate the point in the real space.” Newly amended claim 1 recites this limitation, and therefore claim 4 is of improper dependent form for failing to further limit the subject matter of the claim upon which it depends.
Claim 5 recites “the representation of the atoms of the chemical elements in the unit cell based on the mapping of the crystal structure of the chemical elements comprises training vectors for the machine learning model.” Newly amended claim 1 recites this limitation, and therefore claim 5 is of improper dependent form for failing to further limit the subject matter of the claim upon which it depends.
Claim 10 recites “map the unit cell in the crystal structure to periodic boundary conditions.” Newly amended claim 9 recites this limitation, and therefore claim 10 is of improper dependent form for failing to further limit the subject matter of the claim upon which it depends.
Claim 11 recites “encode the periodic boundary conditions as a Gaussian process.” Newly amended claim 9 recites this limitation, and therefore claim 11 is of improper dependent form for failing to further limit the subject matter of the claim upon which it depends.
Claim 12 recites “interpolate the 3D embedding of the chemical elements proximate the point in the real space.” Newly amended claim 9 recites this limitation, and therefore claim 12 is of improper dependent form for failing to further limit the subject matter of the claim upon which it depends.
Claim 13 recites “the representation of the atoms of the chemical elements in the unit cell being based on the mapping of the crystal structure of the chemical elements comprises training vectors for the machine learning model.” Newly amended claim 9 recites this limitation, and therefore claim 13 is of improper dependent form for failing to further limit the subject matter of the claim upon which it depends.
Claim 18 recites “map the unit cell in the crystal structure to periodic boundary conditions, and to encode the periodic boundary conditions as a Gaussian process.” Newly amended claim 17 recites this limitation, and therefore claim 18 is of improper dependent form for failing to further limit the subject matter of the claim upon which it depends.
Applicant may cancel the claim(s), amend the claim(s) to place the claim(s) in proper dependent form, rewrite the claim(s) in independent form, or present a sufficient showing that the dependent claim(s) complies with the statutory requirements.
Claim Rejections - 35 USC § 101
The Applicant’s amendment received 20 July 2026 has been fully considered, however after further consideration, the rejection of claims 1-20 in the Office action mailed 06 May 2026 has been maintained with modification in view of the amendment, as noted below.
The rejection has been modified to incorporate the newly amended limitations.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claims recite: (a) mathematical concepts, (e.g., mathematical relationships, formulas or equations, mathematical calculations); and (b) mental processes, i.e., concepts performed in the human mind, (e.g., observation, evaluation, judgement, opinion).
Subject matter eligibility evaluation in accordance with MPEP 2106.
Eligibility Step 1: Step 1 of the eligibility analysis asks: Is the claim to a process, machine, manufacture or composition of matter?
Claims 1-8 are directed to a method for representing atomic structures as Gaussian processes (i.e., a process); claims 9-16 are directed to a non-transitory computer-readable medium (i.e., a machine or manufacture); and claim 17-20 are directed to a system comprising a neural processing unit and a memory (i.e., a machine or manufacture).
Therefore, these claims are encompassed by the categories of statutory subject matter, and thus, satisfy the subject matter eligibility requirements under step 1.
[Step 1: YES]
Eligibility Step 2A: First it is determined in Prong One whether a claim recites a judicial exception, and if so, then it is determined in Prong Two whether the recited judicial exception is integrated into a practical application of that exception.
Eligibility Step 2A Prong One: In determining whether a claim is directed to a judicial exception, examination is performed that analyzes whether the claim recites a judicial exception, i.e., whether a law of nature, natural phenomenon, or abstract idea is set forth or described in the claim.
Independent claim 1 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
mapping a crystal structure of chemical elements in a real space, in which atoms of the chemical elements are represented in a unit cell, wherein the mapping comprises mapping the unit cell in the crystal structure to periodic boundary conditions, and encoding the periodic boundary conditions as a Gaussian process (i.e., mathematical concepts);
learning, by a machine learning model, a 3D embedding of each of the chemical elements in the real space according to the mapping of the crystal structure of the chemical elements and the encoded periodic boundary conditions (i.e., mathematical concepts);
training the machine learning model according to a representation of the atoms of the chemical elements in the unit cell based on the mapping of the crystal structure of the chemical elements, wherein the representation comprises training vectors for the machine learning model, and wherein a training set for the machine learning model comprises a vector for each of the atoms in the unit cell based on respective positions and identities (i.e., mathematical concepts); and
predicting a material property corresponding to a point within the real space by interpolating the 3D embedding of the chemical elements proximate the point in the real space (i.e., mathematical concepts).
Independent claim 9 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
program code to map a crystal structure of chemical elements in a real space, in which atoms of the chemical elements are represented in a unit cell, wherein the mapping comprises mapping the unit cell in the crystal structure to periodic boundary conditions, and encoding the periodic boundary conditions as a Gaussian process (i.e., mathematical concepts);
program code to learn, by a machine learning model, a 3D embedding of each of the chemical elements in the real space according to the mapping of the crystal structure of the chemical elements and the encoded periodic boundary conditions (i.e., mathematical concepts);
program code to train the machine learning model according to a representation of the atoms of the chemical elements in the unit cell based on the mapping of the crystal structure of the chemical elements, wherein the representation comprises training vectors for the machine learning model, and wherein a training set for the machine learning model comprises a vector for each of the atoms in the unit cell based on respective positions and identities (i.e., mathematical concepts); and
program code to predict a material property corresponding to a point within the real space by interpolating the 3D embedding of the chemical elements proximate the point in the real space (i.e., mathematical concepts).
Independent claim 17 recites the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas:
map a crystal structure of chemical elements in a real space, in which atoms of the chemical elements are represented in a unit cell, wherein the mapping comprises mapping the unit cell in the crystal structure to periodic boundary conditions, and encoding the periodic boundary conditions as a Gaussian process (i.e., mathematical concepts);
learn, by a machine learning model, a 3D embedding of each of the chemical elements in the real space according to the mapping of the crystal structure of the chemical elements and the encoded periodic boundary conditions (i.e., mathematical concepts);
train the machine learning model according to a representation of the atoms of the chemical elements in the unit cell based on the mapping of the crystal structure of the chemical elements, wherein the representation comprises training vectors for the machine learning model, and wherein a training set for the machine learning model comprises a vector for each of the atoms in the unit cell based on respective positions and identities (i.e., mathematical concepts); and
predict a material property corresponding to a point within the real space by interpolating the 3D embedding of the chemical elements proximate the point in the real space (i.e., mathematical concepts).
Dependent claims 2-8, 10-16, and 18-20 further recite the following steps which fall within the mental processes and/or mathematical concepts groupings of abstract ideas, as noted below.
Dependent claim 2 further recites:
mapping comprises mapping the unit cell in the crystal structure to periodic boundary conditions (i.e., mathematical concepts).
Dependent claim 3 further recites:
encoding the periodic boundary conditions as a Gaussian process (i.e., mathematical concepts).
Dependent claim 4 further recites:
predicting the material property comprises interpolating the 3D embedding of the chemical elements proximate the point in the real space (i.e., mathematical concepts).
Dependent claim 5 further recites:
the representation of the atoms of the chemical elements in the unit cell based on the mapping of the crystal structure of the chemical elements comprises training vectors for the machine learning model (i.e., mathematical concepts).
Dependent claim 6 further recites:
predicting comprises estimating an energy density and/or an electron density as the material property of the point (i.e., mathematical concepts).
Dependent claim 7 further recites:
learning properties and/or a high dimensional representation of the chemical elements using the real space (i.e., mathematical concepts).
Dependent claim 8 further recites:
a training set for the machine learning model comprises a vector for each of the atoms in the unit cell based on respective coordinate positions and identities (X, Y, Z) (i.e., mental processes and mathematical concepts).
Dependent claim 10 further recites:
the program code to map comprises program code to map the unit cell in the crystal structure to periodic boundary conditions (i.e., mental processes and mathematical concepts).
Dependent claim 11 further recites:
program code to encode the periodic boundary conditions as a Gaussian process (i.e., mathematical concepts).
Dependent claim 12 further recites:
the program code to predict the material property comprises program code to interpolate the 3D embedding of the chemical elements proximate the point in the real space (i.e., mathematical concepts).
Dependent claim 13 further recites:
the representation of the atoms of the chemical elements in the unit cell being based on the mapping of the crystal structure of the chemical elements comprises training vectors for the machine learning model (i.e., mathematical concepts).
Dependent claim 14 further recites:
the program code to predict comprises program code to estimate an energy density and/or an electron density as the material property of the point (i.e., mathematical concepts).
Dependent claim 15 further recites:
program code to learn properties and/or a high dimensional representation of the chemical elements using the real space (i.e., mathematical concepts).
Dependent claim 16 further recites:
a training set for the machine learning model comprises a vector for each of the atoms in the unit cell based on respective coordinate positions and identities (X, Y, Z) (i.e., mental processes and mathematical concepts).
Dependent claim 18 further recites:
map the unit cell in the crystal structure to periodic boundary conditions, and to encode the periodic boundary conditions as a Gaussian process (i.e., mathematical concepts).
Dependent claim 19 further recites:
learn properties and/or a high dimensional representation of the chemical elements using the real space (i.e., mathematical concepts).
Dependent claim 20 further recites:
a training set for the machine learning model comprises a vector for each of the atoms in the unit cell based on respective coordinate positions and identities (X, Y, Z) (i.e., mental processes and mathematical concepts).
The abstract ideas recited in the claims are evaluated under the broadest reasonable interpretation (BRI) of the claim limitations when read in light of and consistent with the specification. As noted in the foregoing section, the claims are determined to contain limitations that can practically be performed in the human mind with the aid of a pen and paper (e.g., a training set for the machine learning model comprises a vector for each of the atoms in the unit cell based on respective coordinate positions and identities (X, Y, Z) (claim 8)), and therefore recite judicial exceptions from the mental process grouping of abstract ideas. Additionally, the recited limitations that are identified as judicial exceptions from the mathematical concepts grouping of abstract ideas (e.g., training the machine learning model according to a representation of the atoms of the chemical elements in the unit cell based on the mapping of the crystal structure of the chemical elements) are abstract ideas irrespective of whether or not the limitations are practical to perform in the human mind.
Therefore, claims 1-20 recite an abstract idea.
[Step 2A Prong One: YES]
Eligibility Step 2A Prong Two: In determining whether a claim is directed to a judicial exception, further examination is performed that analyzes if the claim recites additional elements that when examined as a whole integrates the judicial exception(s) into a practical application (MPEP 2106.04(d)). A claim that integrates a judicial exception into a practical application will apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception. The claimed additional elements are analyzed to determine if the abstract idea is integrated into a practical application (MPEP 2106.04(d)(I); MPEP 2106.05(a-h)). If the claim contains no additional elements beyond the abstract idea, the claim fails to integrate the abstract idea into a practical application (MPEP 2106.04(d)(III)).
The judicial exceptions identified in Eligibility Step 2A Prong One are not integrated into a practical application because of the reasons noted below.
In the instant application, the independent claims provide additional elements (e.g., a processor) to perform machine learning mathematical calculations, e.g., training a machine learning model and predicting a material property, however, the claims do not recite any limitations (i.e., additional elements) to which the result of the machine learning model’s prediction, i.e., ‘a material property’ (i.e., the result of the data analysis) is practically applied.
Independent claim 1 and dependent claims 2-8, 10-16, and 18-20 do not recite any elements in addition to the judicial exception, and thus are part of the judicial exception.
The additional elements in independent claim 9 include:
a non-transitory computer-readable medium having program code recorded thereon; and
a processor.
The additional elements in independent claim 17 include:
a neural processing unit; and
a memory coupled to the neural processing unit, and instructions stored in the memory.
The additional elements of a processor (claim 9); a non-transitory computer-readable medium having program code recorded thereon (claim 9); a neural processing unit (claim 17); and a memory coupled to the neural processing unit, and instructions stored in the memory (claim 17); invoke a computer and/or computer-related components merely as tools for use in the claimed process, and therefore are not an improvement to computer functionality itself, or an improvement to any other technology or technical field, and thus, do not integrate the judicial exceptions into a practical application (MPEP 2106.04(d)(1)).
Thus, the additionally recited elements merely invoke a computer and/or computer related components as tools; and as such, when all limitations in claims 1-20 have been considered as a whole (i.e., the analysis takes into consideration all the claim limitations and how those limitations interact and impact each other when evaluating whether the exception is integrated into a practical application), the claims are deemed to not recite any additional elements that would integrate a judicial exception into a practical application, and therefore claims 1-20 are directed to an abstract idea (MPEP 2106.04(d)).
[Step 2A Prong Two: NO]
Eligibility Step 2B: Because the claims recite an abstract idea, and do not integrate that abstract idea into a practical application, the claims are probed for a specific inventive concept. The judicial exception alone cannot provide that inventive concept or practical application (MPEP 2106.05). Identifying whether the additional elements beyond the abstract idea amount to such an inventive concept requires considering the additional elements individually and in combination to determine if they amount to significantly more than the judicial exception (MPEP 2106.05A i-vi).
The claims do not include any additional elements that are sufficient to amount to significantly more than the judicial exception(s) because of the reasons noted below.
Independent claim 1 and dependent claims 2-8, 10-16, and 18-20 do not recite any elements in addition to the judicial exception(s).
The additional elements recited in independent claims 9 and 17 are identified above, and carried over from Step 2A Prong Two along with their conclusions for analysis at Step 2B. Any additional element or combination of elements that was considered to be insignificant extra-solution activity at Step 2A Prong Two was re-evaluated at Step 2B, because if such re-evaluation finds that the element is unconventional or otherwise more than what is well-understood, routine, conventional activity in the field, this finding may indicate that the additional element is no longer considered to be insignificant; and all additional elements and combination of elements were evaluated to determine whether any additional elements or combination of elements are other than what is well-understood, routine, conventional activity in the field, or simply append well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception, per MPEP 2106.05(d).
The additional elements of a processor (claim 9); and a non-transitory computer-readable medium having program code recorded thereon (claim 9); are conventional computer components and/or functions (see MPEP at 2106.05(b) and 2106.05(d)(II) regarding conventionality of computer components and computer processes).
The additional elements of a neural processing unit (claim 17); and a memory coupled to the neural processing unit (claim 17); are conventional. Evidence of conventionality is shown by:
Verhelst et al. (“Machine Learning at the Edge.” (2020). In: Murmann, B., Hoefflinger, B. (eds) NANO-CHIPS 2030. The Frontiers Collection. Springer, Cham. https://doi.org/10.1007/978-3-030-18338-7_18. pp. 293-322, as cited in the Office action mailed 06 May 2026).
Verhelst et al. reviews machine learning at the edge (i.e., data collection devices (e.g., smart watches and smartphones) that contain multiple integrated sensors are often denoted as “edge devices” (page 293, paras. 1-3)), focusing on neural networks as the main machine learning model (page 295, para. 1), and describes the progression from CPU to GPU to NPU as the main neural network inference platforms (page 300, paras. 2-3) and further discusses the Neural Processing Unit (NPU) class of processor, which is a specialized processor that is optimized towards neural network inference in resource constrained devices (page 301, para. 1).
Therefore, when taken alone (i.e., individually), all additional elements in claims 1-20 do not amount to significantly more than the above-identified judicial exception(s). Even when evaluated as an ordered combination, the additional elements fail to transform the exception(s) into a patent-eligible application of that exception. Thus, claims 1-20 are deemed to not contribute an inventive concept, i.e., amount to significantly more than the judicial exception(s) (MPEP 2106.05(II)).
[Step 2B: NO]
Response to Arguments
The Applicant’s arguments/remarks received 20 July 2026 have been fully considered, but are not persuasive.
Regarding Step 2A Prong One, the Applicant states on page 8 (as filed) of the Remarks that the Applicant traverses the rejection of claims 1-20 under 35 U.S.C. 101, and provides the limitations of amended claim 1 (para. 3) and further states (para. 4) that these recitations, when read as a whole, describe a specific machine-learning-based process for encoding crystal structures as Gaussian processes and predicting material properties at arbitrary points in real space via interpolation. The Applicant further states that this process is not practically performable in the human mind, nor does it organize human activity or recite a bare mathematical formula. The Applicant further states that the claimed method operates on crystal-structure data through a defined pipeline of encoding periodic boundary conditions as a Gaussian process, training a machine learning model with per-atom vectors based on positions and identities, and interpolating 3D embeddings to predict material properties at specific points, and accordingly, the claims do not recite a judicial exception, and the eligibility analysis should conclude here.
These arguments/remarks are not persuasive, because first, in Prong One examiners evaluate whether the claim recites a judicial exception, i.e. whether a law of nature, natural phenomenon, or abstract idea is set forth or described in the claim, and as noted in the above rejection, the claims are determined to recite at least one judicial exception, e.g., claim 1 recites “training the machine learning model …” which has been identified as a mathematical concept, and therefore the eligibility analysis does not conclude at Prong One, but rather, must proceed to Prong Two in order to determine whether the recited judicial exception(s) are integrated into a practical application. Second, not all limitations are identified as reciting mental processes, e.g., the limitation reciting “training the machine learning model …” has not been determined to recite a mental process. However, certain limitations have been identified as reciting a mental process, e.g., “a training set for the machine learning model comprises a vector for each of the atoms in the unit cell based on respective coordinate positions and identities (X, Y, Z)” because a crystal structure of a chemical element could consist of a very few, or even a single, atom (e.g., a single molecule of helium consists of one atom) and creating a data vector for such chemical elements would be within the realm of the human mind with the aid of pen and paper.
Regarding Step 2A Prong Two, the Applicant states on page 9 (para. 2) of the Remarks that even assuming arguendo that the claims recite a judicial exception, the additional elements integrate any alleged exception into a practical application, and further states that the claims impose meaningful limits beyond drafting efforts to monopolize an abstract idea. The Applicant further states that specifically, amended claim 1 recites: (1) encoding periodic boundary conditions of a crystal unit cell as a Gaussian process – a specific data transformation that converts crystallographic periodicity into a machine-learnable representation; (2) training a machine learning model using training vectors comprising a vector for each atom in the unit cell based on respective positions and identities – a particular training methodology tied to the physical structure of the material; and (3) predicting a material property corresponding to a point within the real space by interpolating the 3D embedding of the chemical elements proximate the point – a specific prediction technique that enables property determination at arbitrary spatial locations, not merely at known atom positions. The Applicant further states on page 9 (bottom) that these additional elements, individually and in ordered combination, apply the alleged exception in a manner that transforms crystal-structure data into a continuous real-space representation from which material properties can be predicted at any spatial point. The Applicant further states that this constitutes a practical application in the field of computational materials science, and further states that the claims do not merely recite ‘apply it on a computer’ or add insignificant extra-solution activity, but rather, the claims recite a specific technical process that produces a specific technical result: a Gaussian-process-based, spatially continuous representation of a crystal structure that enables pointwise material property prediction via interpolation.
These arguments/remarks are not persuasive, because first, the Applicant’s arguments appear to identify judicial exceptions as additional elements, e.g., the limitations identified at numbers (1), (2), and (3) in the foregoing argument are actually judicial exceptions and not additional elements. Second, a claim that integrates a judicial exception into a practical application will apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception (MPEP 2106.04(d)). However, as noted and discussed in the above rejection, the independent claims provide additional elements (e.g., a processor) to perform machine learning mathematical calculations, e.g., training a machine learning model and predicting a material property, however, the claims do not recite any limitations (i.e., additional elements) to which the result of the machine learning model’s prediction, i.e., ‘a material property’ (i.e., the result of the data analysis) is practically applied, and therefore the claims are not integrated into a practical application at Step 2A Prong Two.
Regarding Step 2B, the Applicant states on page 10 (para. 1) of the Remarks that even assuming arguendo that Step 2A yields an affirmative finding, the claims recite significantly more than any alleged abstract idea. The Applicant further states that the additional elements, viewed as an ordered combination, are not well-understood, routine, or conventional in the field, and further states that the Office action provides no evidentiary support that encoding periodic boundary conditions of a crystal unit cell as a Gaussian process, training a machine learning model with per-atom training vectors based on positions and identities within the unit cell, and predicting material properties at arbitrary real-space points by interpolating 3D embeddings of proximate chemical elements are well-understood, routine, or conventional. The Applicant further states that under the 2019 PEG, examiners should assess whether additional elements are well-understood, routine, or conventional at Step 2B and supply substantial evidence if so asserted. The Applicant further states (para. 2) that the ordered combination of (1) GP encoding of periodic boundary conditions, (2) per-atom training vectors, and (3) pointwise property prediction via GP interpolation of element embeddings reflects an unconventional arrangement that yields a specific technical improvement – enabling continuous, spatially resolved material property prediction from discrete atomic training data. The Applicant further states that this is not generic computer use; it is a specific computational methodology for materials representation and prediction.
These arguments/remarks are not persuasive, because first, the Applicant’s arguments appear to identify judicial exceptions as additional elements, e.g., the limitations identified at numbers (1), (2), and (3) in the foregoing argument are actually judicial exceptions and not additional elements. Second, only the additional elements identified at Step 2A Prong Two are carried over to Step 2B for further evaluation, i.e., a conclusion of whether a claim is eligible at Step 2B requires that all relevant considerations be evaluated, which comprises steps of: (1) carrying over the identification of any additional element(s) in the claim from Step 2A Prong Two; (2) carrying over the conclusions from Step 2A Prong Two on the considerations discussed in MPEP §§ 2106.05(a) - (c), (e) (f) and (h); (3) re-evaluating any additional element or combination of elements that was considered to be insignificant extra-solution activity per MPEP § 2106.05(g), because if such re-evaluation finds that the element is unconventional or otherwise more than what is well-understood, routine, conventional activity in the field, this finding may indicate that the additional element is no longer considered to be insignificant; and (4) evaluating whether any additional element or combination of elements are other than what is well-understood, routine, conventional activity in the field, or simply append well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception, per MPEP § 2106.05(d). As noted in the rejection above, when all additional elements in claims 1-20 have been evaluated individually and as an ordered combination at Step 2B, they are deemed to not contribute an inventive concept, i.e., do not amount to significantly more than the judicial exceptions (MPEP 2106.05(II)). Therefore, the instant claimed technical improvement of enabling continuous, spatially resolved material property prediction from discrete atomic training data, is a purported improvement to the abstract idea (data analysis), and not an improvement to computer functionality itself, or an improvement to another technology or technical field.
Claim Rejections - 35 USC § 103
The Applicant’s amendment received 20 July 2026 has been fully considered, however after further consideration, the rejections under 35 U.S.C. 103 in the Office action mailed 06 May 2026 have been maintained with modification in view of the amendment, as noted below.
The rejection of claims 1-16 under 35 U.S.C. 103 as being unpatentable over Deringer et al. and Li et al. and Pun et al. in the Office action mailed 06 May 2026 has been maintained with modification in view of the amendment, as noted below.
The rejection has been modified to incorporate newly amended limitations.
The rejection of claims 17-20 under 35 U.S.C. 103 as being unpatentable over Deringer et al. and Li et al. and Pun et al. as applied to claims 1-16 above, and further in view of Hwang et al. in the Office action mailed 06 May 2026 has been maintained with modification in view of the amendment, as noted below.
The rejection has been modified to incorporate newly amended limitations.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
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 1-16 are rejected under 35 U.S.C. 103 as being unpatentable over Deringer et al. (“Gaussian Process Regression for Materials and Molecules.” Chemical Reviews, 2021, vol. 121, pp. 10073-10141, as cited in the Office action mailed 06 May 2026) and Li et al. (“Encoding the atomic structure for machine learning in materials science.” WIREs Computational Molecular Science. 2022 (First published 18 June 2021); vol. 12, issue 1; e1558, pp. 1-32, as cited in the Office action mailed 06 May 2026) and Pun et al. (“Physically informed artificial neural networks for atomistic modeling of materials.” Nature Communications, 2019, vol. 10:2339, pp. 1-10, as cited in the Office action mailed 06 May 2026).
Independent claims 1 and 9 encompass a method of using machine learning to predict properties of materials by representing atomic structures as Gaussian processes.
Dependent claims 2-8 and 10-16 further define aspects of the method steps, e.g., encoding the periodic boundary conditions as a Gaussian process, interpolating the 3D embedding of the chemical elements, creating training vectors for the machine learning model, and estimating an energy density and/or an electron density as the material property.
Deringer et al. teaches Gaussian process regression machine-learning methods in computational materials science and chemistry.
Li et al. teaches the encoding of atomic structures for machine learning in materials science, with particular attention to structures of crystalline materials using a crystal unit cell.
Pun et al. teaches the use of a physics-based artificial neural network performing atomistic modeling of materials.
Regarding independent claims 1 and 9, Deringer et al. shows the application of Gaussian process regression to computational chemistry, and the “learning” of general atomistic properties that are of interest for chemical and materials research (page 10074, col. 2, para. 4; and Figure 1); Gaussian process regression (page 10074, Section 2.); interpolation and regression techniques provide tools to fill in the space between data points, resulting in a continuous function representation (page 10075, col. 1, para. 3); the application of machine learning for predicting material properties (page 10119, col. 1, Section 6.5., paras. 1-5); machine-learned (i.e., trained) atomic energies in amorphous silicon as obtained by Gaussian process regression models (page 10116, Figure 33); training multiple models (page 10109, col. 1, para. 2; and Figure 27); and including periodic boundary conditions in calculations (page 10121, col. 2, para. 1).
Regarding independent claims 1 and 9, Deringer et al. does not explicitly show using data of a crystal structure for mapping atoms to a unit cell; or learning a 3D embedding of the chemical elements in the real space according to the mapping of the crystal structure of the chemical elements; or explicitly predicting material properties corresponding to a point within real space.
Regarding independent claims 1 and 9, Li et al. shows mapping a crystal structure and a unit cell (page 4, para. 1, and page 18, para. 1); imposing a periodic boundary condition (page 15, para. 2); and for real space descriptors, the information of lattice vectors, atomic coordinates, and atomic properties such as group number, electronegativity, and first ionization energy, was extracted to construct a feature matrix representing atomic coordinates (page 19, para. 1; and Figure 9(b)).
Regarding independent claims 1 and 9, Pun et al. shows using data of properties of materials that are represented as crystal structures for use in a physics-based neural network model (page 3, col. 2, para. 3); the network is trained by minimizing the error between the energies predicted by the neural network and the respective structure data total energies for a large set of atomic configurations (page 3, col. 1, para. 2); and properties of Aluminum that were predicted using the physics-based neural network model (Table 2).
Regarding dependent claims 2-8 and 10-16, Deringer et al. further shows using periodic boundary conditions (page 10121, col. 2, para. 1) (claims 2, 3, 10, & 11); the Gaussian process regression is a nonlinear, nonparametric regression tool, useful for interpolating between data points scattered in a high-dimensional input space (page 10075, col. 1, para. 4) (claims 4 & 12); training the model using vectors (Figure 27) (claims 5 & 13); electron energy density can be used as an intermediate “descriptor” (page 10098, col. 1, bottom) (claims 6 & 14,).
Regarding dependent claims 7, 8, 15, and 16, Deringer et al. does not show learning properties and/or a high dimensional representation of the chemical elements using the real space; or a training set for the machine learning model comprises a vector for each of the atoms in the unit cell based on respective positions and identities (X, Y, Z).
Regarding dependent claims 7, 8, 15, and 16, Li et al. shows the use of real-space representations as the crystal descriptor in a neural network framework allowing for the discovery of several new materials with targeted properties (page 19).
Therefore, it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method shown by Deringer et al. by incorporating methods for encoding atomic structures for use by machine learning models for predicting materials with target properties, as shown by Li et al. and discussed above. One of ordinary skill in the art would have been motivated to combine the methods of Deringer et al. with the methods of Li et al., because Li et al. shows methods for encoding chemical structures into descriptors that are understandable for a computer. This modification would have had a reasonable expectation of success given that both Deringer et al. and Li et al. disclose methods for using machine learning, and particularly neural networks, for predicting material properties.
It would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method shown by Deringer et al. by incorporating methods for physics-based atomistic modeling of materials, as shown by Pun et al. and discussed above. One of ordinary skill in the art would have been motivated to combine the methods of Deringer et al. with the methods of Pun et al., because Pun et al. shows methods for using a neural network to model the physical nature of interatomic bonding. This modification would have had a reasonable expectation of success given that both Deringer et al. and Pun et al. disclose methods for using machine learning, and particularly neural networks, for predicting material properties.
Claims 17-20 are rejected under 35 U.S.C. 103 as being unpatentable over Deringer et al. and Li et al. and Pun et al. as applied to claims 1-16 above, and further in view of Hwang et al. (“Symbiosis of Semiconductors, AI and Quantum Computing.” IEEE International Electron Devices Meeting (IEDM), San Francisco, CA, USA, 2020, pp. 19.1.1-19.1.6, as cited in the Office action mailed 06 May 2026).
Independent claim 17 encompasses a system for using machine learning to predict properties of materials by representing atomic structures as Gaussian processes.
Dependent claims 18-20 further define characteristics of the systems instructions for representing atomic structures as Gaussian processes.
Hwang et al. teaches various types of AI computing and device technology developments, including the development of AI accelerators, e.g., NPUs: neural processing units) that have become a standard computing unit.
Regarding independent claim 17, Deringer et al. and Li et al. and Pun et al. as applied to claims 1-16 above, do not show a neural processing unit or a memory coupled to the neural processing unit.
Regarding dependent claim 18, Deringer et al. further shows using periodic boundary conditions (page 10121, col. 2, para. 1) (claim 18).
Regarding dependent claims 19 and 20, Li et al. further shows the use of real-space representations as the crystal descriptor in a neural network framework allowing for the discovery of several new materials with targeted properties (page 19).
Regarding independent claim 17, Hwang et al. discusses AI and robotics for material and semiconductor development (page 19.1.2, col. 1, para. 3) and shows the evolution of processors to support AI computation (Figure 2), and further shows that while the shift from CPU to GPU occurred because the GPU architecture is well fitted to parallel computations, NPU are fully dedicated to AI compute with simpler architectures with lower precisions and massive ALUs (i.e., arithmetic logic unit) (page 19.1.1, col. 1, para. 1).
Therefore, it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method shown by Deringer et al. and Li et al. and Pun et al. as applied to claims 1-16 above, by incorporating a neural processing unit, as shown by Hwang et al. and discussed above. One of ordinary skill in the art would have been motivated to combine the methods of Deringer et al. and Li et al. and Pun et al. as applied to claims 1-16 above, with the methods of Hwang et al. because Hwang et al. shows that NPUs are fully dedicated to AI computing. This modification would have had a reasonable expectation of success given that both Deringer et al. and Li et al. and Pun et al. as applied to claims 1-16 above, and Hwang et al. disclose AI computing and material development.
Response to Arguments
The Applicant’s arguments/remarks received 20 July 2026 have been fully considered, but are not persuasive.
The Applicant states on page 11 (top) of the Remarks that the Applicant traverses the rejections under 35 U.S.C. 103. The Applicant further states that Deringer, Li, and Pun were cited against claim 1, and further states that as noted in the Office action, Deringer does not disclose: (1) using crystal-structure data for mapping atoms to a unit cell; (2) learning a 3D embedding of chemical elements in real space according to the mapping; and (3) explicitly predicting material properties corresponding to a point in real space. The Applicant further states that in particular, Deringer discusses GP regression generally but does not teach encoding periodic boundary conditions of a crystal unit cell as a GP for real-space property prediction. The Applicant further states that Li involves general crystal structure/unit cell concepts but does not teach the claimed GP-based real-space embedding approach, and further states that in particular, Li describes generic ‘descriptors’ which review various descriptor types but does not teach a specific GP-based, per atom training vector approach. The Applicant further states that Pun is directed to neural networks for structure-level property prediction, not pointwise real-space property prediction via GP interpolation, and further states that in particular, Pun uses neural networks for structure-level property prediction, not pointwise real-space property prediction via GP interpolation of element embeddings. The Applicant further states on page 11 (bottom) that the combination of Deringer, Li, and Pun fails to teach or suggest the limitations as in claims 1, 9, and 17, and further states that Hwang, which is directed to AI computing and device technology developments addressing memory bottlenecks through near-memory, fails to rectify the noted deficiencies of Deringer, Li, and Pun, and further states on page 12 that accordingly, independent claims 1, 9, and 17 are believed to be in condition for allowance.
These arguments/remarks are not persuasive, because first, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986). Second, As noted and discussed in the above rejections, Deringer et al. teaches Gaussian process regression machine-learning methods in computational materials science and chemistry; Li et al. teaches the encoding of atomic structures for machine learning in materials science, with particular attention to structures of crystalline materials using a crystal unit cell; Pun et al. teaches the use of a physics-based artificial neural network performing atomistic modeling of materials; and Hwang et al. teaches various types of AI computing and device technology developments, including the development of AI accelerators, e.g., NPUs: neural processing units) that have become a standard computing unit. Thus, the combination of references used in the above rejections show that the claimed invention would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, for the reasons provided in the foregoing motivation statements.
Conclusion
No claims are allowed.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Inquiries
Any inquiry concerning this communication or earlier communications from the examiner should be directed to STEVEN W. BAILEY whose telephone number is (571)272-8170. The examiner can normally be reached Mon - Fri. 1000 - 1800.
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, KARLHEINZ SKOWRONEK can be reached at (571) 272-9047. 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.
/S.W.B./Examiner, Art Unit 1687
/Karlheinz R. Skowronek/Supervisory Patent Examiner, Art Unit 1687
