DETAILED ACTION
Status of Claims
This Office action is responsive to communications filed on 2024-05-23. Claim(s) 1-10 and 14-23 is/are pending and are examined herein.
Claim(s) 23 invoke(s) 35 USC 112(f).
Claim(s) 1-10 and 14-23 is/are rejected under 35 USC 112(b).
Claim(s) 23 is/are rejected under 35 USC 112(a).
Claim(s) 1-10 and 14-23 is/are rejected under 35 USC 101.
Claim(s) 1-10 and 14-23 is/are rejected under 35 USC 103.
Notice of Pre-AIA or AIA Status
The present application, filed on or after 2013-03-16, is being examined under the first inventor to file provisions of the AIA .
Priority
The present application claims priority from EP application 21214040.4, filed on 13 December 2021-12-13. Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Information Disclosure Statement
The attached information disclosure statement(s) (IDS), submitted on 2024-05-23, is/are in compliance with the provisions of 37 CFR 1.97. Accordingly, the attached information disclosure statement(s) is/are being considered by the examiner.
Examiner’s Remarks
Claims 2-6 and 15-19 recite a number of modules of various apparatuses. These uses of the generic placeholder “module” do not currently invoke interpretation under 35 USC 112(f) because the claims presently attribute no specific functionality to these modules themselves. The applicant is advised that, if the claims were amended to attribute specific functionality to these modules without clarifying the specific structures that would accomplish those functionalities, these claim elements would then invoke interpretation under 112(f).
Claim Interpretation – 35 USC 112(f)
The following is a quotation of 35 USC 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 USC 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph, is invoked.
As explained in MPEP 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph:
the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph, except as otherwise indicated in an Office action.
Claim(s) 23 recite(s) at least one limitation that invoke an interpretation under 35 USC 112(f). It recites a data storage device storing program instructions operative, when performed by the one or more processors, to cause the one or more processors to perform at least the method of claim 1. This includes the generic placeholder “device” alongside the functional language “storing program instructions operative, when performed by the one or more processors, to cause the one or more processors to perform at least the method of claim 1”. Moreover, the claim does not describe any explicit structure for a “data storage device” that is sufficient for performing the claimed functionality, and the specification merely recites identical language [specification, 0025] without providing any clear indication of the structure that this claim element is to have. For the purpose of compact prosecution, the data storage device is interpreted to be a machine-readable medium as described in [specification, 0109].
If applicant does not intend to have this/these limitation(s) interpreted under 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph, applicant may:
amend the claim limitation(s) to avoid it/them being interpreted under 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or
present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 USC 112(f) or pre-AIA 35 USC 112, sixth paragraph.
Claim Rejections - 35 USC 112(b)
The following is a quotation of 35 USC 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 USC 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim(s) 1-10 and 14-23 is/are rejected under 35 USC 112(b) or 35 USC 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 USC 112, the applicant), regards as the invention.
Claims 1 and 14 include at least the following instances of indefinite language:
They recite the first data set comprises a plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a plurality of features derived from second sensor data measured from one or more other production apparatuses [emphasis added] but this repeat nomenclature results in ambiguous antecedent basis. The examiner suggests “the first data set comprises a first plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a second plurality of features derived from second sensor data measured from one or more other production apparatuses” to avoid this issue (and to provide clear antecedent basis for subsequent recitations of the features comprised in these data sets that appear in dependent claims, as suggested below).
They recite a classification value indicative of whether the received data item is from the first or second data sets [emphasis added] but the underlined phrase lacks antecedent basis. The examiner suggests “a classification value indicative of whether the received data item is from the first data set or the second data set” for clarity of antecedent basis.
They recite inputting a second subset of the first data set and/or a second subset of the second data set to the trained machine learning model [emphasis added] but the use of “and/or” renders unclear the intended scope of the claim unclear. The examiner suggests either “inputting a second subset of the first data set
They recite evaluating the accuracy of the classification values [emphasis added] but the underlined phrases lack antecedent basis. The examiner suggests “evaluating an accuracy of the corresponding classification values” for proper antecedent basis.
They recite using the evaluated accuracy of the classification values [emphasis added] but the underlined phrase lacks antecedent basis. The examiner suggests “the corresponding classification values” for proper antecedent basis.
They recite determining the presence of a fault in the subject production apparatus [emphasis added] but the former underlined phrase lacks antecedent basis and the second is repeated nomenclature (since the claim previously recites “diagnosing a fault in a subject production apparatus”). The examiner suggests “determining a presence of the fault in the subject production apparatus” for clarity of antecedent basis.
Dependent claims 2-10 and 15-23 inherit these rejections.
Claims 2 and 15 include at least the following instances of indefinite language:
They recite each of the subject production apparatus and the one or more other production apparatuses comprises a plurality of corresponding modules [emphasis added] but it is not clear whether the correspondence is to be between the modules of a single apparatus, or between the sets of modules of different apparatuses. The language of this limitation on its face suggests the former interpretation, though the subsequent limitation recited in these claims (cf. below) suggests the latter. MPEP 2173.05(b) indicates that a “claim may be rendered indefinite when a limitation of the claim is defined by reference to an object and the relationship between the limitation and the object is not sufficiently defined” and, in the present instance, the relationship between the apparatuses and modules is not sufficiently defined. The examiner suggests either removing the word “corresponding” (i.e., “each of the subject production apparatus and the one or more other production apparatuses comprises a plurality of
They recite the first and second subsets of both the first and second data sets comprise features derived from measurements made respectively of a first module of the subject production apparatus and corresponding module of the one or more other production apparatuses [emphasis added]. The first two underlined phrases lack antecedent basis, and the last underlined phrase is ungrammatical due to a missing article. While those issues may have straightforward fixes, there would nonetheless remain two more substantial issues that would require more substantial amendments to clarify the intended scope of this entire limitation. First, the use of the word “features” results in ambiguous nomenclature: the parent claim recites both “a [first] plurality of features derived from first sensor data” and “a [second] plurality of features derived from second sensor data” which renders unclear whether the “features” appearing in this dependent claim are one of those two pluralities, or some other features besides those. Secondly, the limitation does not make clear what pair of sets the word “respectively” is intended to operate over. More precisely, it is not clear whether the measurements from the “first module” are intended to be comprised in the first data set, or the first subset of the first data set, or the first subset of the second data set, or the second subset of the first data set, or the second subset of the second data set; and similarly, it is not clear which of these five sets the measurements from the “corresponding module” are comprised in. As best understood by the examiner in view of the specification, the intention of this limitation may have been that “the first plurality of features includes a first feature derived from first measurements of a first module of the subject production apparatus, and the second plurality of features includes a second feature derived from second measurements of a second module of the one or more other production apparatuses” (making use of the amendments suggested under the 112(b) rejections of the parent claim). For the purpose of compact prosecution, the limitation is interpreted broadly as encompassing at least this interpretation.
Dependent claims 3-6 and 16-19 inherit the rejections.
Claims 3 and 16 include at least the following instances of indefinite language:
They recite the plurality of features of the first dataset comprise statistical aggregations of measurements of the first module of the subject production apparatus, and the plurality of features of the second dataset comprise statistical aggregations of measurements of the corresponding module of the one or more other production apparatuses [emphasis added] but the underlined phrases either have ambiguous antecedent basis, lack antecedent basis, or are instances of repeated nomenclature. In view of the suggested amendments to the parent claims, the examiner suggests “the first plurality of features s first statistical aggregations of the first measurements second plurality of features s second statistical aggregations of the second measurements
Similarly, claims 4 and 17 include at least the following instances of indefinite language:
They recite the plurality of features of the first dataset comprise raw sensor data from measurements of the first module of the subject production apparatus, and the plurality of features of the second dataset comprise raw sensor data from measurements of the corresponding module of the one or more other production apparatuses [emphasis added] but the underlined phrases either have ambiguous antecedent basis, lack antecedent basis, or are instances of repeated nomenclature. In view of the suggested amendments to the parent claims, the examiner suggests “the first plurality of features s first raw sensor data from the first measurements second plurality of features s second raw sensor data from the second measurements
Claims 5 and 18 include at least the following instances of indefinite language:
They recite relating to a corresponding module but this is repeated nomenclature since their respective parent claims already recite “a corresponding module”, rendering unclear whether the “corresponding module” of this claim is the same as, or different from, that recited in the parent claim. As best understood by the examiner, the intended antecedent appears to be the earlier recitation of “each of one or more other modules” appearing in the same limitation, and the examiner correspondingly suggests “relating to the other module” to clarify this antecedent. For the purpose of compact prosecution, the claim is interpreted as encompassing at least this interpretation.
They recite using the corresponding additional first data set and additional second data set but this phrase lacks antecedent basis. The examiner suggests “using the respective additional first data set and the respective additional second data set” for proper antecedent basis.
Claim 5 recites the performinq, inputtinq, evaluatinq and usinq steps using the corresponding additional first data set and additional second data set but the underlined lacks antecedent basis and the intended antecedent is not clear. In particular, the only limitations beginning with “performing”, “inputting”, “evaluating”, and “using” appear to be those recited in the parent claim, but those limitations use the “first data set” and the “second data set”, not the “respective additional first data set” and the “respective additional second data set”, so it is not clear if those steps in the parent claim are the intended antecedent. Even if they are, the language used in the claim does not make clear what role the “respective additional first data set” and the “respective additional second data set” are to play in those limitations. MPEP 2173.05(b) indicates that a “claim may be rendered indefinite when a limitation of the claim is defined by reference to an object and the relationship between the limitation and the object is not sufficiently defined” and, in the present instance, the relationship between the respective additional first/second data sets and these limitations is insufficiently defined. Alternative language clarifying this relationship is advised.
Claim 18 recites perform the performance of the supervise training, input of the second subset, evaluation of the accuracy and use of the evaluated accuracy using the corresponding additional first data set and additional second data set [emphasis added] but the first underlined phrase lacks antecedent basis, and the remaining are ungrammatical. As noted above, even if this language is intended to refer back to limitations recited in the parent claim, the language used in the claim does not make clear what role the “respective additional first data set” and the “respective additional second data set” are to play in those limitations. Alternative language consistent with the language used in claim 5 is advised.
Dependent claims 6 and 19 inherit the rejections.
Claims 6 and 19 include at least the following instances of indefinite language:
They recite a subset of the modules of the subject production apparatus [emphasis added] but this has ambiguous antecedent basis since the parent claims introduce both a “plurality of modules” and “one or more other modules”. The examiner notes that the subsequent recitation of respective evaluation accuracies suggests that the intended antecedent may have been the latter, and the examiner corresponding suggests amending to “a subset of the one or more other modules of the subject production apparatus” for proper antecedent basis. For the purpose of compact prosecution, the claim is interpreted as such herein.
Claim 19 recites the diagnostic process comprises identification of a subset… and provision of an indication [emphasis added] but the underlying phrases are ungrammatical. The examiner suggests “the diagnostic process comprises identifying a subset… and providing an indication” for grammaticality and for maintaining parallel language with claim 6.
Claims 7 and 20 include at least the following instances of indefinite language:
They recite the supervised machine learning algorithm but this lacks antecedent basis. The examiner suggests “the machine learning model” for proper antecedent basis (noting that this does not change the scope of the claim since the parent claim already indicates that “supervised training” is performed on the “machine learning model”).
They recite each of a plurality of features of the first and second datasets [emphasis added] but the parent claims already introduce pluralities of features. In view of the amendments regarding the parent claims above, the examiner suggests “each feature of the first plurality of features and the second plurality of features
They recite the importance of the feature [emphasis added] but this lacks antecedent basis. This should be “an importance of the feature” to avoid this issue.
They recite determining the classification value but this has ambiguous antecedent basis since the parent claim recites both “output a classification value” and “obtain corresponding classification values”, rendering unclear the intended antecedent of the claim. As best understood by the examiner in view of the specification, the examiner suggests modifying the dependent to read “determining the corresponding classification values” to avoid this issue. For the purpose of compact prosecution, the claim is interpreted as such herein.
They recite the importance parameter but this lacks antecedent basis. This should be “the respective importance parameter” for proper antecedent basis.
Claim 20 recites the diagnostic process comprises use of the supervised machine learning algorithm… and identification of at least one of the features… [emphasis added] but the underlying phrases are ungrammatical. The examiner suggests “the diagnostic process comprises using the machine learning model… and identifying at least one of the features” for grammaticality and for maintaining parallel language with claim 7.
Dependent claims 8-9 and 21 inherit these rejections.
Claims 8, 9 and 21 include at least the following instances of indefinite language:
They recite wherein the importance parameter represents [emphasis added] (once each in claims 8 and 9, and twice in claim 21) but the underlined phrase lacks antecedent basis. In keeping with the suggestions made above, the examiner suggests “wherein the respective importance parameter of each feature of the first plurality of features and the second plurality of features represents” for proper antecedent basis.
Claims 9 and 21 recite the ability of a feature to decrease a class impurity [emphasis added] but the former phrase lacks antecedent basis and the latter is repeated nomenclature (since the parent claim already recites a “plurality of features”). The examiner suggests “an ability of the feature to decrease a class impurity” for proper antecedent basis (where the antecedent of “the feature” would be provided by the phrase “each feature” as suggested above).
Claims 10 and 22 include at least the following instances of indefinite language:
They recite the production apparatus but this lacks antecedent basis. It should be “the subject production apparatus” for proper antecedent basis.
Claim 23 recites an element invoking interpretation under 35 USC 112(f). However, the specification provides no clear interpretation as to how to interpret this claim element. Moreover, even if this element is interpreted as referring to the memory or hard drive of a generic computer, MPEP 2181(II)(B) indicates that “the structure be more than simply a general purpose computer or microprocessor and that the specification must disclose an algorithm for performing the claimed function”, that “[a]n algorithm is defined, for example, as ‘a finite sequence of steps for solving a logical or mathematical problem or performing a task’” and that “a rejection under 35 USC 112(b) or pre-AIA 35 USC 112, second paragraph is appropriate if the specification discloses no corresponding algorithm associated with a computer or microprocessor”. In the present instance, the originally filed specification provides no explicit algorithms for the various steps of parent claim 1, so claim 23 is rejected for failing to disclose structure sufficient for performing the claimed functionalities.
Claim Rejections - 35 USC 112(a)
The following is a quotation of the first paragraph of 35 USC 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 USC 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claim(s) 23 is/are rejected under 35 USC 112(a) or 35 USC 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 USC 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Claim(s) 23 recite(s) limitations invoked 35 USC 112(f) and are rejected under 35 USC 112(b) for failing to disclose sufficient structure. MPEP 2181(II)(B) indicates that “[w]hen a claim containing a computer-implemented 35 USC 112(f) claim limitation is found to be indefinite under 35 USC 112(b) for failure to disclose sufficient corresponding structure (e.g., the computer and the algorithm) in the specification that performs the entire claimed function, it will also lack written description under 35 USC 112(a)”. Consequently, claim(s) 23 is/are rejected under 35 USC 112(a) for lack of written description.
Claim Rejections - 35 USC 101 - Statutory Categories
35 USC 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim(s) 14-23 is/are rejected under 35 USC 101 because the claimed invention is directed to non-statutory subject matter. The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter for the following reasons.
Claim 14 is directed to a computer program product comprising a computer-readable storage medium comprising instructions therein which, when executed by a computer system, are configured to cause the computer system to perform certain steps [emphasis added]. MPEP 2106.03 indicates that “transitory forms of signal transmission” are not a statutory category. Moreover, the specification does not include an explicit disavowal of transitory signals; in fact, the specification explicitly indicates that a “machine-readable medium… may include electrical, optical, acoustical or other forms of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc)” [specification, 0109]. Consequently, the claim is rejected for not falling within the four categories of patent eligible subject matter. Dependent claims 15-22 inherit the rejection. The examiner suggests amending the claim to “… a non-transitory computer-readable storage medium…” to avoid this issue.
Claim 23 recites a data storage device. As noted above, this claim element invokes interpretation under 35 USC 112(f), but the specification provides no clear structure to this claim element. In particular, the specification does not include an explicit disavowal of transitory signals. The examiner suggests amending to “a non-transitory computer-readable storage medium
Claim Rejections - 35 USC 101 - Abstract Idea
35 USC 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim(s) 1-10 and 14-23 is/are rejected under 35 USC 101 because the claimed invention(s) is/are directed to abstract ideas without significantly more.
Claim 1
Step 1. The claim and its dependents 2-10 and 23 fall under the statutory category of methods. An analysis of step 2 for each of these claims follows.
Step 2A Prong 1. The claim recites the following abstract ideas:
A method of diagnosing a fault in a subject production apparatus, the method comprising: (This recites a mental process that can be performed in the human mind, since a human mind can diagnose faults in machines. See MPEP 2106.04(a)(2)(III).)
obtain corresponding classification values; (This recites a mental process that can be performed in the human mind, since a human mind can obtain classification values. See MPEP 2106.04(a)(2)(III).)
evaluating the accuracy of the classification values; (This recites a mathematical concept and/or a mental process that can be performed in the human mind or by a human using pen and paper, since a human mind can evaluate the accuracy of classification values. See MPEP 2106.04(a)(2)(I, III).)
and using the evaluated accuracy of the classification values in performing a diagnostic process for determining the presence of a fault in the subject production apparatus. (This recites a mental process that can be performed in the human mind, since a human mind can perform a diagnostic process for detecting a fault. See MPEP 2106.04(a)(2)(III).)
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
obtaining a first data set and a second data set, (This recites insignificant extra-solution activity. See MPEP 2106.05(g).)
wherein the first data set comprises a plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a plurality of features derived from second sensor data measured from one or more other production apparatuses; (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
performing supervised training of a machine learning model (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
using a first subset of the first data set and a first subset of the second data set, (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
wherein the machine learning model is trained, upon receiving a data item selectively from the first or second data sets, to [output a classification value indicative of whether the received data item is from the first or second data sets;] (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
output a classification value indicative of whether the received data item is from the first or second data sets; (This recites insignificant extra-solution activity. See MPEP 2106.05(g).)
inputting a second subset of the first data set and/or a second subset of the second data set to the trained machine learning model to [obtain corresponding classification values] (This recites insignificant extra-solution activity. See MPEP 2106.05(g).)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
obtaining a first data set and a second data set, (This insignificant extra-solution activity is well-understood, routine, conventional as it is mere data transfer. See MPEP 2106.05(d)(II), “Receiving or transmitting data over a network” and/or “Storing and retrieving information in memory”.)
wherein the first data set comprises a plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a plurality of features derived from second sensor data measured from one or more other production apparatuses; (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
performing supervised training of a machine learning model (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
using a first subset of the first data set and a first subset of the second data set, (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
wherein the machine learning model is trained, upon receiving a data item selectively from the first or second data sets, to [output a classification value indicative of whether the received data item is from the first or second data sets;] (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
output a classification value indicative of whether the received data item is from the first or second data sets; (The insignificant extra-solution activity is well-understood, routine, conventional as it is merely presenting output. See MPEP 2106.05(d)(II), “Presenting offers”.)
inputting a second subset of the first data set and/or a second subset of the second data set to the trained machine learning model to [obtain corresponding classification values] (This insignificant extra-solution activity is well-understood, routine, conventional as it is mere data transfer. See MPEP 2106.05(d)(II), “Receiving or transmitting data over a network” and/or “Storing and retrieving information in memory”.)
Claim 2
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
[The method according to claim 1, wherein] each of the subject production apparatus and the one or more other production apparatuses comprises a plurality of corresponding modules, (This recites a general link between an abstract idea and a particular field of use or technological environment. See MPEP 2106.05(h).)
and the first and second subsets of both the first and second data sets comprise features derived from measurements made respectively of a first module of the subject production apparatus and corresponding module of the one or more other production apparatuses. (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
[The method according to claim 1, wherein] each of the subject production apparatus and the one or more other production apparatuses comprises a plurality of corresponding modules, (This recites a general link between an abstract idea and a particular field of use or technological environment. See MPEP 2106.05(h).)
and the first and second subsets of both the first and second data sets comprise features derived from measurements made respectively of a first module of the subject production apparatus and corresponding module of the one or more other production apparatuses. (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
Claim 3
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
[The method according to claim 2, wherein] the plurality of features of the first dataset comprise statistical aggregations of measurements of the first module of the subject production apparatus, and the plurality of features of the second dataset comprise statistical aggregations of measurements of the corresponding module of the one or more other production apparatuses. (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
[The method according to claim 2, wherein] the plurality of features of the first dataset comprise statistical aggregations of measurements of the first module of the subject production apparatus, and the plurality of features of the second dataset comprise statistical aggregations of measurements of the corresponding module of the one or more other production apparatuses. (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
Claim 4
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
[The method according to claim 2, wherein] the plurality of features of the first dataset comprise raw sensor data from measurements of the first module of the subject production apparatus, and the plurality of features of the second dataset comprise raw sensor data from measurements of the corresponding module of the one or more other production apparatuses. (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
[The method according to claim 2, wherein] the plurality of features of the first dataset comprise raw sensor data from measurements of the first module of the subject production apparatus, and the plurality of features of the second dataset comprise raw sensor data from measurements of the corresponding module of the one or more other production apparatuses. (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
Claim 5
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
and, for each of the other modules of the subject production apparatus, performing the performinq, inputtinq, evaluatinq and usinq steps using the corresponding additional first data set and additional second data set, and a corresponding machine learning model, (This includes recitations of the same mental processes as indicated in the parent claim.)
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
[The method according to claim 2, further comprising,] for each of one or more other modules of the subject production apparatus, obtaining a respective additional first data set relating to that module of the subject production apparatus and a respective additional second data set relating to a corresponding module of the one or more other production apparatuses, (This recites insignificant extra-solution activity. See MPEP 2106.05(g).)
and, for each of the other modules of the subject production apparatus, performing the performinq, inputtinq, evaluatinq and usinq steps using the corresponding additional first data set and additional second data set, and a corresponding machine learning model, (This includes recitations of the same additional elements as indicated in the parent claim.)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
[The method according to claim 2, further comprising,] for each of one or more other modules of the subject production apparatus, obtaining a respective additional first data set relating to that module of the subject production apparatus and a respective additional second data set relating to a corresponding module of the one or more other production apparatuses, (This recites insignificant extra-solution activity. See MPEP 2106.05(g).)
and, for each of the other modules of the subject production apparatus, performing the performinq, inputtinq, evaluatinq and usinq steps using the corresponding additional first data set and additional second data set, and a corresponding machine learning model, (This includes recitations of the same additional elements as indicated in the parent claim.)
Claim 6
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
[The method according to claim 5, in which the diagnostic process comprises] identifying a subset of the modules of the subject production apparatus for which the respective evaluation accuracies are highest, (This recites a mental process that can be performed in the human mind or by a human using pen and paper. See MPEP 2106.04(a)(2)(III).)
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
and providing an indication of the identified subset of modules to a user. (This recites insignificant extra-solution activity. See MPEP 2106.05(g).)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
and providing an indication of the identified subset of modules to a user. (The insignificant extra-solution activity is well-understood, routine, conventional as it is merely presenting output. See MPEP 2106.05(d)(II), “Presenting offers”.)
Claim 7
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
determine from each of a plurality of features of the first and second datasets a respective importance parameter indicative of the importance of the feature in determining the classification value, (This recites a mathematical concept and a mental process that can be performed in the human mind or by a human using pen and paper. See MPEP 2106.04(a)(2)(I, III).)
and identifying at least one of the features for which the importance parameter is highest. (This recites a mental process that can be performed in the human mind or by a human using pen and paper. See MPEP 2106.04(a)(2)(III).)
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
[The method according to claim 1, wherein the diagnostic process comprises] using the supervised machine learning algorithm to (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
[The method according to claim 1, wherein the diagnostic process comprises] using the supervised machine learning algorithm to (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
Claim 8
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
[The method according to claim 7, wherein] the machine learning model is a logistic regression classifier, (This recites a mathematical concept. See MPEP 2106.04(a)(2)(I).)
and wherein the importance parameter represents a magnitude of a feature coefficient of the logistic regression classifier. (This recites a mathematical concept. See MPEP 2106.04(a)(2)(I).)
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
Claim 9
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
[The method according to claim 7, wherein] the machine learning model is a random forest classifier, (This recites a mathematical concept. See MPEP 2106.04(a)(2)(I).)
and wherein the importance parameter represents the ability of a feature to decrease a class impurity. (This recites a mathematical concept. See MPEP 2106.04(a)(2)(I).)
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
Claim 10
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
[The method according to claim 1, wherein] the production apparatus is a lithographic apparatus. (This recites a general link between an abstract idea and a particular field of use or technological environment. See MPEP 2106.05(h).)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
[The method according to claim 1, wherein] the production apparatus is a lithographic apparatus. (This recites a general link between an abstract idea and a particular field of use or technological environment. See MPEP 2106.05(h).)
Claim 23
Step 2A Prong 1. The claim recites the following abstract ideas:
The abstract idea(s) in the parent claim(s).
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
The additional element(s) in the parent claim(s).
A computer system comprising: one or more processors; and a data storage device storing program instructions operative, when performed by the one or more processors, to cause the one or more processors to perform at least [the method of claim 1.] (This recites generic computing components for performing an abstract idea. See MPEP 2106.05(f)(2).)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
The additional element(s) in the parent claim(s).
A computer system comprising: one or more processors; and a data storage device storing program instructions operative, when performed by the one or more processors, to cause the one or more processors to perform at least [the method of claim 1.] (This recites generic computing components for performing an abstract idea. See MPEP 2106.05(f)(2).)
Claim 14
Step 2A Prong 1. The claim recites the following abstract ideas:
obtain corresponding classification values; (This recites a mental process that can be performed in the human mind, since a human mind can obtain classification values. See MPEP 2106.04(a)(2)(III).)
evaluate the accuracy of the classification values; (This recites a mathematical concept and/or a mental process that can be performed in the human mind or by a human using pen and paper, since a human mind can evaluate the accuracy of classification values. See MPEP 2106.04(a)(2)(I, III).)
and use the evaluated accuracy of the classification values in performance of a diagnostic process for determination of the presence of a fault in the subject production apparatus. (This recites a mental process that can be performed in the human mind, since a human mind can perform a diagnostic process for detecting a fault. See MPEP 2106.04(a)(2)(III).)
Step 2A Prong 2. The claim recites the following additional elements which, considered individually and as an ordered combination, do not integrate the abstract idea into a practical application:
A computer program product comprising a computer-readable storage medium comprising instructions therein which, when executed by a computer system, are configured to cause the computer system to at least: (This recites generic computing components for performing an abstract idea. See MPEP 2106.05(f)(2).)
obtain a first data set and a second data set, (This recites insignificant extra-solution activity. See MPEP 2106.05(g).)
wherein the first data set comprises a plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a plurality of features derived from second sensor data measured from one or more other production apparatuses; (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
perform supervised training of a machine learning model (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
using a first subset of the first data set and a first subset of the second data set, (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
wherein the machine learning model is trained, upon receiving a data item selectively from the first or second data sets, to [output a classification value indicative of whether the received data item is from the first or second data sets;] (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
output a classification value indicative of whether the received data item is from the first or second data sets; (This recites insignificant extra-solution activity. See MPEP 2106.05(g).)
input a second subset of the first data set and/or a second subset of the second data set to the trained machine learning model to [obtain corresponding classification values] (This recites insignificant extra-solution activity. See MPEP 2106.05(g).)
Step 2B. The claim recites the following additional elements which, considered individually and as an ordered combination, do not amount to significantly more than the abstract idea:
A computer program product comprising a computer-readable storage medium comprising instructions therein which, when executed by a computer system, are configured to cause the computer system to at least: (This recites generic computing components for performing an abstract idea. See MPEP 2106.05(f)(2).)
obtain a first data set and a second data set, (This insignificant extra-solution activity is well-understood, routine, conventional as it is mere data transfer. See MPEP 2106.05(d)(II), “Receiving or transmitting data over a network” and/or “Storing and retrieving information in memory”.)
wherein the first data set comprises a plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a plurality of features derived from second sensor data measured from one or more other production apparatuses; (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
perform supervised training of a machine learning model (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
using a first subset of the first data set and a first subset of the second data set, (This recites data of a particular type or source, merely linking an abstract idea to a particular field of use. See MPEP 2106.05(h).)
wherein the machine learning model is trained, upon receiving a data item selectively from the first or second data sets, to [output a classification value indicative of whether the received data item is from the first or second data sets;] (This recites merely applying (or equivalent) an abstract idea, or implementing an abstract idea on a computer, or using a computer as a tool to perform an abstract idea. See MPEP 2106.05(f).)
output a classification value indicative of whether the received data item is from the first or second data sets; (The insignificant extra-solution activity is well-understood, routine, conventional as it is merely presenting output. See MPEP 2106.05(d)(II), “Presenting offers”.)
input a second subset of the first data set and/or a second subset of the second data set to the trained machine learning model to [obtain corresponding classification values] (This insignificant extra-solution activity is well-understood, routine, conventional as it is mere data transfer. See MPEP 2106.05(d)(II), “Receiving or transmitting data over a network” and/or “Storing and retrieving information in memory”.)
Claims 15-22 inherit limitations from claim 14 and recite additional limitations which are substantially similar to those recited by claims 2-8 and 10, respectively, so they are rejected by the same rationale. (The examiner notes that claim 21 also includes the limitations of claim 9 in the alternative, so claim 21 can also be rejected by the rationale described therein.)
Claim Rejections - 35 USC 103
The following is a quotation of 35 USC 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.
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 USC 102(b)(2)(C) for any potential 35 USC 102(a)(2) prior art against the later invention.
Claim(s) 1-2, 4-5, 10, 14-15, 17-18, and 22-23 is/are rejected under 35 USC 103 as being unpatentable over David LOPEZ-PAZ et al. (Revisiting Classifier Two-Sample Tests, published 2018; hereafter, “Lopez”) in view of Chao ZHANG et al. (Auto-Encoder Based Model for High Dimensional Imbalanced Industrial Data, published 2021-12-02; hereafter, “Zhang”).
Claim 1
Lopez discloses:
obtaining a first data set and a second data set, ([Lopez, sections 2-3]: Lopez discusses classifier two-sample testing (C2ST), i.e., a method for two-sample testing using binary classifiers. The inputs of this method include “two samples S_P and S_Q” [Lopez, section 3], where S_P = {x_1, …, x_n} and S_Q = {y_1, …, y_m} [Lopez, section 2], the examples x_i and y_j being drawn from a set X [Lopez, section 3]. The sets S_P and S_Q map respectively to the “first data set” and the “second data set” of the claim. The applicant is invited to consult Friedman as cited in the conclusion of this Office action for another disclosure of C2ST.)
performing supervised training of a machine learning model using a first subset of the first data set and a first subset of the second data set, wherein the machine learning model is trained, upon receiving a data item selectively from the first or second data sets, to output a classification value indicative of whether the received data item is from the first or second data sets; ([Lopez, section 3]: Lopez discusses assigning a label of 0 to the examples x_i in S_P and a label of 1 to the examples y_j in S_Q, constructing the union dataset D = {(x_i, 0) : i = 1, …, n} ∪ {(y_j, 1) : j = 1, …, m} ⊆ X × {0, 1}, splitting D into disjoint training and testing subsets D_{tr} and D_{te}, respectively, and training a binary classifier f : X to [0, 1] on D_{tr} [Lopez, section 3]. The examples in S_P that are in D_{tr} map to the “first subset of the first data set” of the claim, and the examples in S_Q that in D_{tr} map to the “first subset of the second data set” of the claim. The binary classifier f trained on D_{tr} maps to the “machine learning model” of the claim: its output is “indicative of whether the received data item is from the first or second data sets” as recited by the claim, and its training is “supervised” since the training data D_{tr} includes the labels 0 or 1.)
inputting a second subset of the first data set and/or a second subset of the second data set to the trained machine learning model to obtain corresponding classification values; evaluating the accuracy of the classification values; ([Lopez, section 3]: Lopez discusses computing a “classification accuracy on D_{te}” denoted hat{t} (also called the “C2ST statistic”) [Lopez, section 3]. The examples in S_P that are in D_{te} map to the “second subset of the first data set” and the examples in S_Q that are in D_{te} map to the “second subset of the second data set” of the claim. The values f(z_i) for (z_i, l_i) in D_{te} map to the “corresponding classification values” of the claim, and the classification accuracy hat{t} maps to the “accuracy of the classification values” of the claim.)
and using the evaluated accuracy of the classification values in performing a diagnostic process ([Lopez, section 3]: Lopez discusses “comput[ing] a p-value using the null distribution of the C2ST” hat{t} in order to “accept or reject the null hypothesis” that the two samples S_P and S_Q are drawn from the same distribution [Lopez, section 3]. The computation of this p-value to test this null hypothesis maps to the “diagnostic process” of the claim.)
While Lopez discusses a variety of applications of C2ST [Lopez, section 4], it might not specifically discuss an application to fault detection in an industrial setting. In other words, Lopez might not distinctly disclose:
A method of diagnosing a fault in a subject production apparatus, the method comprising:
[obtaining a first data set and a second data set,] wherein the first data set comprises a plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a plurality of features derived from second sensor data measured from one or more other production apparatuses;
[performing a diagnostic process] for determining the presence of a fault in the subject production apparatus.
Zhang is in the field of machine learning. Moreover, Lopez in view of Zhang discloses:
A method of diagnosing a fault in a subject production apparatus, the method comprising: ([Zhang, sections 1, and 2.1, and 3]: Zhang makes use of “wafer manufacturing data from Seagate” [Zhang, section 1 last paragraph] which includes “14 physical deposition equipment”, each of which “is a vacuum tool that deposits thin film” [Zhang, section 3 paragraph beginning “The tool family”]. The “main objective is to predict whether a wafer passes measurements” [Zhang, section 2.1 paragraph beginning “We simplified”]. Any given vacuum tool from this set of 14 maps to the “subject production apparatus” of the claim, and a wafer failing to pass measurements maps to the “fault” of the claim.)
[obtaining a first data set and a second data set,] wherein the first data set comprises a plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a plurality of features derived from second sensor data measured from one or more other production apparatuses; ([Zhang, section 3]: Zhang discloses that each of the 14 tools “has 130 sensors installed” which capture data about process variables [Zhang, section 3]. As noted above, one of these 14 tools maps to the “subject production apparatus” of the claim. The data from this tool maps to the “first sensor data measured from the subject production apparatus” of the claim, and the process variables in this data map to one of the “[first] plurality of features” of the claim. In the combination with Lopez, this data plays the role of S_P from Lopez, i.e., it serves as the “first data set” of the claim as mapped above. Any nonempty subset of the other 13 tools map to the “one or more other production apparatuses” of the claim, the data from those tools to the “second sensor data measured from [the] one or more other production apparatuses”, and the process variables in this data to the “[second] plurality of features” of the claim. In the combination with Lopez, this data plays the role of S_Q from Lopez, i.e., it serves as the the “second data set” of the claim.)
[performing a diagnostic process] for determining the presence of a fault in the subject production apparatus. ([Zhang, section 2.1]: As noted above, Zhang discloses that their “main objective is to predict whether a wafer passes measurements” [Zhang, section 2.1 paragraph beginning “We simplified”], with a wafer failing to pass mapping to the “fault” of the claim. In the combination, C2ST as discussed in Lopez is used to detect such a fault in the manner described above.)
Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art to use the Classifier Two-Sample Tests as described in Lopez for fault detection in industrial data as described in Zhang because the industrial data sets described in Zhang “are much larger and more complex than public open soft sensor data” and “reflect the complex and noisy nature of real-world data” [Zhang, abstract], “[d]eep learning is emerging as a promising technique to analyze” such data due to its “strong representational capabilities on complex data and the flexibility it offers from an architectural perspective” [Zhang, abstract], and “Classifier Two-Sample Tests (C2ST) learn a suitable representation of the data on the fly, return test statistics in interpretable units, have simple asymptotic distributions, and their learned features and predictive uncertainty provide interpretation on how P and Q differ” [Lopez, section 1 paragraph beginning “Intriguingly”], thereby resulting in an effective fault detection system for industrial applications.
Claim 2
Lopez in view of Zhang discloses the elements of the parent claim(s). It also discloses:
[The method according to claim 1, wherein] each of the subject production apparatus and the one or more other production apparatuses comprises a plurality of corresponding modules, ([Zhang, section 3]: As noted under the parent claim, each of the 14 tools “has 130 sensors installed” [Zhang, section 3]. A sensor falls under the broadest reasonable interpretation of a “module” as recited by the claim. The examiner notes that, while the nature of the correspondence required by this limitation is indefinite (cf. 112(b) rejections), the sensors on a single tool are in correspondence with each other in the since that they are all installed on the same tool, and the sets of sensors on any two of the tools are in correspondence with each other in the sense that there are 130 on each tool so there is a bijection (i.e., a one-to-one correspondence) between these sets, so there is a “correspondence” in either of two possible interpretations of the limitation as best understood by the examiner. The applicant is also invited to consult David as cited below, which discusses “many components on the lithography apparatus” [David, 0066] and/or “one or more components of the lithographic apparatus” [David, 0090].)
and the first and second subsets of both the first and second data sets comprise features derived from measurements made respectively of a first module of the subject production apparatus and corresponding module of the one or more other production apparatuses. ([Lopez, section 3; Zhang, section 3]: Any given sensor on the tool mapped to the “subject production apparatus” in the parent claim maps to the “first module of the subject production apparatus” of the claim. Moreover, any given sensor on any other tool maps to the “corresponding module of the one or more other production apparatuses” of the claim; it is “corresponding” in the sense, for example, that it is part of the same industrial process. Then the features in the first data set as mapped under the parent claim include data from the “first module” as mapped herein, and the features in the second data set as mapped under the parent claim include data from the “corresponding module” as mapped herein, so the mappings fall under the broadest reasonable interpretation of this limitation as best understood by the examiner in view of the 112(b) rejections.)
The same motivation to combine applies.
Claim 4
Lopez in view of Zhang discloses the elements of the parent claims. It also discloses:
[The method according to claim 2, wherein] the plurality of features of the first dataset comprise raw sensor data from measurements of the first module of the subject production apparatus, and the plurality of features of the second dataset comprise raw sensor data from measurements of the corresponding module of the one or more other production apparatuses. ([Zhang, section 3]: As noted under the parent claim, the variables in the datasets come from the sensors capturing data about process variables [Zhang, section 3]. The variables are thus “raw sensor data” as recited by the claim. The applicant is also invited to consult David as cited in the rejection of claim 3, which indicates that “input data can be upstream metrology data, or data from process equipment” [David, 0046].)
The same motivation to combine applies.
Claim 5
Lopez in view of Zhang discloses the elements of the parent claims. It also discloses:
[The method according to claim 2, further comprising,] for each of one or more other modules of the subject production apparatus, ([Zhang, section 3]: As above, the sensors of the tool that is mapped to the “subject production apparatus” above map to the “one or more other modules” of the claim.) obtaining a respective additional first data set relating to that module of the subject production apparatus ([Zhang, section 3]: The data from a given sensor of the “subject production apparatus” maps to the “respective additional first data set” of the claim.) and a respective additional second data set relating to a corresponding module of the one or more other production apparatuses, ([Zhang, section 3]: As noted under the parent claim, any sensor on any of the other tools maps to the “corresponding module” of the claim. The data from that sensor maps to the “respective additional second data set” of the claim.)
and, for each of the other modules of the subject production apparatus, performing the performinq, inputtinq, evaluatinq and usinq steps using the corresponding additional first data set and additional second data set, and a corresponding machine learning model, to obtain a respective evaluation accuracy for each of the one or more other modules. ([Lopez, section 3; Zhang, section 3]: In the combination, all of the data of Zhang is used in the C2ST methodology described in Lopez. This means, in particular, that the “performing, inputting, evaluating, and using” steps as mapped under the parent claim use, a fortiori, the “respective first additional data set” and the “respective additional second data set” as mapped above. The binary classifier of Lopez maps to the “corresponding machine learning model” of the claim and the C2ST statistic to the “respective evaluation accuracy” of the claim.)
The same motivation to combine applies.
Claim 10
Lopez in view of Zhang discloses the elements of the parent claim(s). It also discloses:
[The method according to claim 1, wherein] the production apparatus is a lithographic apparatus. ([Zhang, sections 1 and 3]: The applicant’s specification defines that a “lithographic apparatus is a machine constructed to apply a desired pattern onto a substrate” [specification, 0003]. The tools in Zhang are “semiconductor process manufacturing tools” [Zhang, section 1] which “deposit[…] thin film” onto a “substrate” [Zhang, section 3 paragraph beginning “The process tool”]. The “[subject] production apparatus” of the claim as mapped under the parent claim is thus a “lithographic apparatus” as required by this claim.)
The same motivation to combine applies.
Claim 23
Lopez in view of Zhang disclose the elements of the parent claim(s). They also disclose:
A computer system comprising: one or more processors; and a data storage device storing program instructions operative, when performed by the one or more processors, to cause the one or more processors to perform at least [the method of claim 1]. ([Lopez, section 4; Zhang, section 2]: Lopez discloses an implementation of the methods disclosed therein that is made available through GitHub [Lopez, section 4 first paragraph]. The implementation maps to the “program instructions” of the claim. Any computer on which this implementation is executed (or, alternatively, the GitHub server on which this implementation is stored) maps to the “computer system” of the claim, its processors mapping to the “one or more processors” and its memory or hard drive mapping to the “data storage device” of the claim. The examiner notes that Zhang also discloses the use of a computer having a “NVIDIA Tesla V100 SXM2 GPU” [Zhang, section 2 last paragraph].)
The same motivation to combine applies.
Claim 14
Lopez discloses:
A computer program product comprising a computer-readable storage medium comprising instructions therein which, when executed by a computer system, are configured to cause the computer system to at least: ([Lopez, section 4; Zhang, section 2]: Lopez discloses an implementation of the methods disclosed therein that is made available through GitHub [Lopez, section 4 first paragraph]. The implementation maps to the “instructions” and the “computer program product” of the claim. Any computer on which this implementation is executed (or, alternatively, the GitHub server on which this implementation is stored) maps to the “computer system” of the claim, its memory or hard drive mapping to the “computer-readable storage medium” of the claim.)
obtain a first data set and a second data set, ([Lopez, sections 2-3]: Lopez discusses classifier two-sample testing (C2ST), i.e., a method for two-sample testing using binary classifiers. The inputs of this method include “two samples S_P and S_Q” [Lopez, section 3], where S_P = {x_1, …, x_n} and S_Q = {y_1, …, y_m} [Lopez, section 2], the examples x_i and y_j being drawn from a set X [Lopez, section 3]. The sets S_P and S_Q map respectively to the “first data set” and the “second data set” of the claim. The applicant is invited to consult Friedman as cited in the conclusion of this Office action for another disclosure of C2ST.)
perform supervised training of a machine learning model using a first subset of the first data set and a first subset of the second data set, wherein the machine learning model is trained, upon receiving a data item selectively from the first or second data sets, to output a classification value indicative of whether the received data item is from the first or second data sets; ([Lopez, section 3]: Lopez discusses assigning a label of 0 to the examples x_i in S_P and a label of 1 to the examples y_j in S_Q, constructing the union dataset D = {(x_i, 0) : i = 1, …, n} ∪ {(y_j, 1) : j = 1, …, m} ⊆ X × {0, 1}, splitting D into disjoint training and testing subsets D_{tr} and D_{te}, respectively, and training a binary classifier f : X to [0, 1] on D_{tr} [Lopez, section 3]. The examples in S_P that are in D_{tr} map to the “first subset of the first data set” of the claim, and the examples in S_Q that in D_{tr} map to the “first subset of the second data set” of the claim. The binary classifier f trained on D_{tr} maps to the “machine learning model” of the claim: its output is “indicative of whether the received data item is from the first or second data sets” as recited by the claim, and its training is “supervised” since the training data D_{tr} includes the labels 0 or 1.)
input a second subset of the first data set and/or a second subset of the second data set to the trained machine learning model to obtain corresponding classification values; evaluating the accuracy of the classification values; ([Lopez, section 3]: Lopez discusses computing a “classification accuracy on D_{te}” denoted hat{t} (also called the “C2ST statistic”) [Lopez, section 3]. The examples in S_P that are in D_{te} map to the “second subset of the first data set” and the examples in S_Q that are in D_{te} map to the “second subset of the second data set” of the claim. The values f(z_i) for (z_i, l_i) in D_{te} map to the “corresponding classification values” of the claim, and the classification accuracy hat{t} maps to the “accuracy of the classification values” of the claim.)
and use the evaluated accuracy of the classification values in performing a diagnostic process ([Lopez, section 3]: Lopez discusses “comput[ing] a p-value using the null distribution of the C2ST” hat{t} in order to “accept or reject the null hypothesis” that the two samples S_P and S_Q are drawn from the same distribution [Lopez, section 3]. The computation of this p-value to test this null hypothesis maps to the “diagnostic process” of the claim.)
While Lopez discusses a variety of applications of C2ST [Lopez, section 4], it might not specifically discuss an application to fault detection in an industrial setting. In other words, Lopez might not distinctly disclose:
A method of diagnosing a fault in a subject production apparatus, the method comprising:
[obtaining a first data set and a second data set,] wherein the first data set comprises a plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a plurality of features derived from second sensor data measured from one or more other production apparatuses;
[performing a diagnostic process] for determining the presence of a fault in the subject production apparatus.
Zhang is in the field of machine learning. Moreover, Lopez in view of Zhang discloses:
A method of diagnosing a fault in a subject production apparatus, the method comprising: ([Zhang, sections 1, and 2.1, and 3]: Zhang makes use of “wafer manufacturing data from Seagate” [Zhang, section 1 last paragraph] which includes “14 physical deposition equipment”, each of which “is a vacuum tool that deposits thin film” [Zhang, section 3 paragraph beginning “The tool family”]. The “main objective is to predict whether a wafer passes measurements” [Zhang, section 2.1 paragraph beginning “We simplified”]. Any given vacuum tool from this set of 14 maps to the “subject production apparatus” of the claim, and a wafer failing to pass measurements maps to the “fault” of the claim.)
[obtaining a first data set and a second data set,] wherein the first data set comprises a plurality of features derived from first sensor data measured from the subject production apparatus, and the second data set comprises a plurality of features derived from second sensor data measured from one or more other production apparatuses; ([Zhang, section 3]: Zhang discloses that each of the 14 tools “has 130 sensors installed” which capture data about process variables [Zhang, section 3]. As noted above, one of these 14 tools maps to the “subject production apparatus” of the claim. The data from this tool maps to the “first sensor data measured from the subject production apparatus” of the claim, and the process variables in this data map to one of the “[first] plurality of features” of the claim. In the combination with Lopez, this data plays the role of S_P from Lopez, i.e., it serves as the “first data set” of the claim as mapped above. Any nonempty subset of the other 13 tools map to the “one or more other production apparatuses” of the claim, the data from those tools to the “second sensor data measured from [the] one or more other production apparatuses”, and the process variables in this data to the “[second] plurality of features” of the claim. In the combination with Lopez, this data plays the role of S_Q from Lopez, i.e., it serves as the the “second data set” of the claim.)
[performing a diagnostic process] for determining the presence of a fault in the subject production apparatus. ([Zhang, section 2.1]: As noted above, Zhang discloses that their “main objective is to predict whether a wafer passes measurements” [Zhang, section 2.1 paragraph beginning “We simplified”], with a wafer failing to pass mapping to the “fault” of the claim. In the combination, C2ST as discussed in Lopez is used to detect such a fault in the manner described above.)
Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art to use the Classifier Two-Sample Tests as described in Lopez for fault detection in industrial data as described in Zhang because the industrial data sets described in Zhang “are much larger and more complex than public open soft sensor data” and “reflect the complex and noisy nature of real-world data” [Zhang, abstract], “[d]eep learning is emerging as a promising technique to analyze” such data due to its “strong representational capabilities on complex data and the flexibility it offers from an architectural perspective” [Zhang, abstract], and “Classifier Two-Sample Tests (C2ST) learn a suitable representation of the data on the fly, return test statistics in interpretable units, have simple asymptotic distributions, and their learned features and predictive uncertainty provide interpretation on how P and Q differ” [Lopez, section 1 paragraph beginning “Intriguingly”], thereby resulting in an effective fault detection system for industrial applications.
Claims 15, 17-18, and 22 inherit limitations from claim 14 and recite additional limitations which are substantially similar to those recited by claims 2, 4-5, and 10, respectively, so they are rejected by the same rationale.
Claim(s) 3 and 16 is/are rejected under 35 USC 103 as being unpatentable over Lopez in view of Zhang, further in view of Jeffrey DAVID (US20170109646A1, published 2017-04-20; hereafter, “David”).
Claim 3
Lopez in view of Zhang discloses the elements of the parent claim(s). It might not distinctly disclose:
[The method according to claim 2, wherein] the plurality of features of the first dataset comprise statistical aggregations of measurements of the first module of the subject production apparatus, and the plurality of features of the second dataset comprise statistical aggregations of measurements of the corresponding module of the one or more other production apparatuses.
David is in the field of machine learning. It in particular discusses the use of machine learning in “semiconductor manufacturing processes” [David, abstract], including for “fault detection” [David, 0051]. Moreover, Lopez in view of Zhang and David discloses:
[The method according to claim 2, wherein] the plurality of features of the first dataset comprise statistical aggregations of measurements of the first module of the subject production apparatus, and the plurality of features of the second dataset comprise statistical aggregations of measurements of the corresponding module of the one or more other production apparatuses. ([David, 0046, 0048]: David indicates that “input data can be upstream metrology data, or data from process equipment” [David, 0046] and further indicates that “[d]ata transformation or feature engineering can be performed in in-situ spectral data or other sensor data that is collected during a particular process” [David, 0048]. It indicates, for example, that “[s]tatistics such as mean, standard deviation, min, and max may be collected at each wavelength interval of the spectral set over time and used as data inputs” [David, 0048]. In the combination, the industrial fault detection method of Lopez in view of Zhang is modified to include these statistical transformations of the sensor data, which then map to the “statistical aggregations” of the claim.)
Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art to combine the industrial fault detection method of Lopez in view of Zhang with the techniques described in David because “better process control and integration schemes are needed now more than ever” [David, 0006] and “incorporating new types of input data” can help “optimiz[e] testing and burn-in procedures” [David, 0029].
Claim 16 recites limitations which are substantially similar to those recited by claim 3, so it is rejected by the same rationale.
Claim(s) 6 and 19 is/are rejected under 35 USC 103 as being unpatentable over Lopez in view of Zhang, further in view of Sumanth NARASAPPA (US20150074035A1, published 2015-03-12; hereafter, “Narasappa”).
Claim 6
Lopez in view of Zhang discloses the elements of the parent claim(s). It might not distinctly disclose:
[The method according to claim 5, in which the diagnostic process comprises] identifying a subset of the modules of the subject production apparatus for which the respective evaluation accuracies are highest,
and providing an indication of the identified subset of modules to a user.
Narasappa is in the field of machine learning. It discloses a method for “[f]inding a root cause” for the failure of a system having “a large set of associated components” [Narasappa, 0002]. In the combination with Lopez and Zhang, the system on which this root cause analysis is performed is the “subject production apparatus” of the claim as mapped above. The components of the system as described in Narasappa then correspond to the “modules” of the claim. Moreover, Lopez in view of Zhang and Narasappa discloses:
[The method according to claim 5, in which the diagnostic process comprises] identifying a subset of the modules of the subject production apparatus for which the respective evaluation accuracies are highest, ([Narasappa, 0018]: Narasappa discloses a method of “root cause analysis” in which “[c]omponents found to have the highest scores during anomaly detection of associated KPIs may be identified as a root cause set” [Narasappa, 0027]. In the combination, the fault detection method of Lopez in view of Zhang is used as the anomaly detection method, so that the classification accuracies of Lopez play the role of the scores obtained during anomaly detection as in Narasappa. Then the root cause set of Narasappa maps to the “subset” of the claim, since it is the subset of components “for which the respective evaluation accuracies are highest” as required by the claim.)
and providing an indication of the identified subset of modules to a user. ([Narasappa, 0018]: Narasappa discloses “output[ting] one or more root cause sets” [Narasappa, 0018]. Outputting falls under the broadest reasonable interpretation of “providing an indication… to a user” as recited by the claim.)
Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art to combine the industrial fault detection method of Lopez in view of Zhang with the methods described in Narasappa because they include “improved techniques for determining root causes” which overcome “shortcomings of existing approaches” [Narasappa, 0005], thereby providing a more effective system overall.
Claim 19 recites limitations which are substantially similar to those recited by claim 6, so it is rejected by the same rationale.
Claim(s) 7-9 and 20-21 is/are rejected under 35 USC 103 as being unpatentable over Lopez in view of Zhang, further in view of Trevor HASTIE et al. (The Elements of Statistical Learning, published 2009; hereafter, “Hastie”).
Claim 7
Lopez in view of Zhang discloses the elements of the parent claim(s). While it discloses the use of different types of classifiers in C2ST (namely, neural networks and k-nearest neighbors [Lopez, section 4 first paragraph]), it might not distinctly disclose:
[The method according to claim 1, wherein] the diagnostic process comprises using the supervised machine learning algorithm to determine from each of a plurality of features of the first and second datasets a respective importance parameter indicative of the importance of the feature in determining the classification value, and identifying at least one of the features for which the importance parameter is highest.
Hastie is in the field of statistical learning. Moreover, Lopez in view of Zhang and Hastie discloses:
[The method according to claim 1, wherein] the diagnostic process comprises using the supervised machine learning algorithm to determine from each of a plurality of features of the first and second datasets a respective importance parameter indicative of the importance of the feature in determining the classification value, and identifying at least one of the features for which the importance parameter is highest. ([Hastie, section 4.4]: Hastie discusses many different types of classifiers, including logistic regression models [Hastie, section 4.4]. In the combination, the classifier of Lopez (i.e., the “machine learning model” and/or “supervised machine learning algorithm” of the claim) can be taken to be a logistic regression model as in Hastie. In the context of logistic regression, Hastie discloses “search[ing] for a parsimonious model involving a subset of the variables” [Hastie, section 4.4.1 last paragraph], giving an example involving heart disease data in which a list of 7 predictor variables is pruned down to a list of 4 by successively removing variables whose coefficients have a “nonsignificant Z-score” (i.e., a Z-score whose absolute value is small) [Hastie, section 4.4.2 and tables 4.2-3]. The absolute value of the Z-score of a predictor variable maps to its “importance parameter” as recited by the claim, and the list of predictor variables used in the final model are then the “at least one of the features for which the importance parameter is highest” as recited by the claim.)
Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art to use logistic regressions as described in Hastie for the industrial fault detection method of Lopez in view of Zhang because Lopez already discusses the use of different types of classifiers in C2ST [Lopez, section 4 first paragraph] and logistic regression models have the advantage that they help in “understand[ing] the role of the input variables in explaining the outcome” [Hastie, section 4.4.1 last paragraph], so the combination would result in a more transparent system.
Claim 8
Lopez in view of Zhang and Hastie discloses the elements of the parent claim(s). It also discloses:
[The method according to claim 7,] wherein the machine learning model is a logistic regression classifier, ([Hastie, section 4.4]: As noted under the parent claim, Hastie discusses the use of logistic regression for classification [Hastie, section 4.4], and, in the combination, the classifier of Lopez can be taken to be a logistic regression model as in Hastie, so that the “machine learning model” of the claim as mapped under the parent claims is a “logistic regression classifier” as required herein.)
and wherein the importance parameter represents a magnitude of a feature coefficient of the logistic regression classifier. ([Hastie, section 4.4]: As noted under the parent claim, the absolute value of the Z-score of a coefficient of a predictor variable was mapped to the “importance parameter” of the claim. The coefficient itself maps to the “feature coefficient” of the claim, and the absolute value of the Z-score “represents a magnitude of [the] feature coefficient” as required by the claim.)
The same motivation to combine applies.
Claim 9
Lopez in view of Zhang and Hastie discloses the elements of the parent claim(s). It also discloses:
[The method according to claim 7, wherein] the machine learning model is a random forest classifier, ([Hastie, chapter 15]: Hastie discusses many different types of classifiers, including random forests [Hastie, chapter 15 and algorithm 15.1]. In the combination, the classifier of Lopez can be taken to be a random forest as in Hastie, so that the “machine learning model” of the claim as mapped under the parent claims is in fact a “random forest classifier” as required herein.)
and wherein the importance parameter represents the ability of a feature to decrease a class impurity. ([Hastie, sections 9.2.3 and 15.3.2]: Hastie discusses constructing variable importance plots for random forests to identify the most important predictor variables [Hastie, section 15.3.2 and figure 15.5; see also, section 10.13]. This includes a measurement of importance based on Gini splitting index [Hastie, figure 15.5 left], which is a measure of “node impurity” [Hastie, section 9.2.3]. In other words, the Gini index of a variable can be mapped to the “importance parameter” of the claim so that it “represents [an] ability of [the] feature to decrease a class impurity” as required by the claim. The applicant is also invited to consult Nembrini as cited in the conclusion of this Office action.)
Before the effective filing date of the invention, it would have been obvious to a person of ordinary skill in the art to use random forests as described in Hastie for the industrial fault detection method of Lopez in view of Zhang because Lopez already discusses the use of different types of classifiers in C2ST [Lopez, section 4 first paragraph] and because random forests “do remarkably well, with very little tuning required” [Hastie, section 15.2 paragraph beginning “The authors”], so the combination would result in a more effective system.
Claims 20-21 recite limitations which are substantially similar to those recited by claim 7-8, respectively, so they are rejected by the same rationale. (The examiner notes that claim 21 also includes the limitations of claim 9 in the alternative, so claim 21 can also be rejected by the rationale described therein.)
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Jerome FRIEDMAN (On Multivariate Goodness-of-Fit and Two-Sample Testing, published 2003; hereafter, “Friedman”) discusses the use of binary classification for two-sample testing [Friedman, sections I and III].
Stefano NEMBRINI et al. (The revival of the Gini importance?, published 2018; hereafter “Nembrini”) discusses random forests and the fact that an “important reason for their popularity is the availability of variable importance measures (VIMs)” such as “mean decrease of impurity (MDI)” as measured by “Gini importance” [Nembrini, first paragraph].
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Shishir AGRAWAL whose telephone number is +1 703-756-1183. The examiner can normally be reached Monday through Thursday, 08:30-14:30 Pacific Time.
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 on +1 571-270-3428. The fax phone number for the organization where this application or proceeding is assigned is +1 571-273-8300.
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 +1 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call +1 800-786-9199 (IN USA OR CANADA) or +1 571-272-1000.
/S.A./Examiner, Art Unit 2123
/ALEXEY SHMATOV/Supervisory Patent Examiner, Art Unit 2123
Prosecution Insights