DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim1-20 rejected under 35 U.S.C. 101 because the claimed invention is directed towards an abstract idea without significantly more.
Subject Matter Eligibility Analysis Step 1:
Claims 1-14 recite method claims. Claims 15-20 are machine/system/product claims.
Therefore, claims 1-20 are directed to one of the four statutory categories of patentable subject matter.
Regarding claim 1:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 1 recites the step:
“identifying current sensor data associated with processing of substrates by substrate processing equipment,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually evaluate the sensor data.)
Thus, claim 1 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 1 recites the additional elements:
“providing the current sensor data as input to a trained machine learning model, the trained machine learning model being trained using training input comprising a first window of time of historical sensor data and target output comprising the first window of time or a second window of time of the historical sensor data to generate the trained machine learning model, the historical sensor data being associated with normal runs of processing of historical substrates by the substrate processing equipment;”(This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes high level recitation of training a machine learning model.)
“obtaining, from the trained machine learning model, one or more outputs;” (This element does not integrate the abstract idea into a practical application because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g))
“and causing, based on the one or more outputs, an anomaly response action associated with the substrate processing equipment.” (This element does not integrate the abstract idea into a practical application because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g))
Therefore, claim 1 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 1 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“providing the current sensor data as input to a trained machine learning model, the trained machine learning model being trained using training input comprising a first window of time of historical sensor data and target output comprising the first window of time or a second window of time of the historical sensor data to generate the trained machine learning model, the historical sensor data being associated with normal runs of processing of historical substrates by the substrate processing equipment;”(This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes high level recitation of training a machine learning model.)
“obtaining, from the trained machine learning model, one or more outputs;” (This element does not provide significantly more because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g)) in the form of storing and retrieving information, which is well-known, routine, and conventional (see MPEP 2106.05(d)(II)).)
“and causing, based on the one or more outputs, an anomaly response action associated with the substrate processing equipment.” (This element does not provide significantly more because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g)) in the form of presenting offers and gathering statistics, which is well-known, routine, and conventional (see MPEP 2106.05(d)(II)).)
Therefore, claim 1 is subject-matter ineligible.
Regarding claim 2:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 2 recites the step:
“The method of claim 1,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 1.)
Thus, claim 2 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 2 recites the additional elements:
“wherein the target output comprises the first window of time of the historical sensor data, the target output being same as the training input.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 2 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 2 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“wherein the target output comprises the first window of time of the historical sensor data, the target output being same as the training input.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 2 is subject-matter ineligible.
Regarding claim 3:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 3 recites the step:
“The method of claim 1,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 1.)
Thus, claim 3 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 3 recites the additional elements:
“wherein the target output comprises the second window of time of the historical sensor data, the target output being offset from the training input by one or more windows of time.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 3 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 3 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“wherein the target output comprises the second window of time of the historical sensor data, the target output being offset from the training input by one or more windows of time.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 3 is subject-matter ineligible.
Regarding claim 4:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 4 recites the step:
“The method of claim 1,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 1.)
Thus, claim 4 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 4 recites the additional elements:
“wherein: the current sensor data comprises a plurality of sequenced data sets at a first set of windows of time;” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
“the one or more outputs comprise reconstruction data comprising predicted sequenced data sets at a second set of windows of time;” (This element does not integrate the abstract idea into a practical application because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g))
“and each window of time of the second set of windows of time is offset from a corresponding window of time of the first set of windows of time by one or more windows of time.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 4 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 4 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“wherein: the current sensor data comprises a plurality of sequenced data sets at a first set of windows of time;” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
“the one or more outputs comprise reconstruction data comprising predicted sequenced data sets at a second set of windows of time;” (This element does not provide significantly more because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g)) in the form of storing and retrieving information, which is well-known, routine, and conventional (see MPEP 2106.05(d)(II)). Examiner notes further limiting of data that is retrieved.)
“and each window of time of the second set of windows of time is offset from a corresponding window of time of the first set of windows of time by one or more windows of time.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 4 is subject-matter ineligible.
Regarding claim 5:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 5 recites the step:
“The method of claim 1,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 1.)
Thus, claim 5 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 5 recites the additional elements:
“wherein the trained machine learning model is a trained long short-term memory (LSTM) recurrent neural network (RNN) model.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes high level recitation of trained model.)
Therefore, claim 5 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 5 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“wherein the trained machine learning model is a trained long short-term memory (LSTM) recurrent neural network (RNN) model.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes high level recitation of trained model.)
Therefore, claim 5 is subject-matter ineligible.
Regarding claim 6:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 6 recites the step:
“The method of claim 5,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 5.)
“wherein the encoder determines a compressed representation of the input,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually determine a compressed representation of the input.)
“wherein the decoder uses the compressed representation to predict a future plurality of sequenced data sets.” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually make a prediction.)
Thus, claim 6 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 6 recites the additional elements:
“wherein the trained LSTM RNN model comprises an encoder and a decoder,” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes claim limitation refers to the high level recitation of training a machine learning model.)
“wherein the input comprises a current plurality of sequenced data sets,” (This element does not integrate the abstract idea into a practical application because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g))
Therefore, claim 6 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 6 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“wherein the trained LSTM RNN model comprises an encoder and a decoder,” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes claim limitation refers to the high level recitation of training a machine learning model.)
“wherein the input comprises a current plurality of sequenced data sets,” (This element does not provide significantly more because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g)) in the form of storing and retrieving information, which is well-known, routine, and conventional (see MPEP 2106.05(d)(II)).)
Therefore, claim 6 is subject-matter ineligible.
Regarding claim 7:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 7 recites the step:
“The method of claim 1, wherein the causing of the anomaly response action comprises:” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 1.)
“comparing the current sensor data to reconstruction data associated with the one or more outputs to generate model reconstruction error;” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually compare data and determine error.)
“and identifying an anomaly responsive to determining that the model reconstruction error is greater than a threshold error.” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually determine error is greater than a threshold.)
Thus, claim 7 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 7 does not further recite any additional elements. Therefore, 7 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
Since there are no additional elements, claim 7 does not provide significantly more than the abstract idea itself, taken alone and in combination.
Therefore, claim 7 is subject-matter ineligible.
Regarding claim 8:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 8 recites the step:
“The method of claim 7 further comprising:” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 7.)
“generating a plurality of anomaly scores based on the one or more outputs, wherein each of the plurality of anomaly scores corresponds to a respective sensor of a plurality of sensors;” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually determine a plurality of anomaly scores with consideration for the outputs.)
“and ranking contribution to the model reconstruction error by each of the plurality of sensors based on the plurality of anomaly scores.” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually rank contribution to the model reconstruction error with consideration for the anomaly scores.)
Thus, claim 8 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 8 does not further recite any additional elements. Therefore, 8 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
Since there are no additional elements, claim 8 does not provide significantly more than the abstract idea itself, taken alone and in combination.
Therefore, claim 8 is subject-matter ineligible.
Regarding claim 9:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 9 recites the step:
“A method comprising: identifying historical sensor data associated with normal runs of processing of historical substrates by substrate processing equipment;” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually identify that historical sensor data is associated with normal runs.)
Thus, claim 9 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 9 recites the additional elements:
“and training a machine learning model using training input comprising a first window of time of the historical sensor data and target output comprising the first window of time or a second window of time of the historical sensor data to generate a trained machine learning model, the trained machine learning model to perform an anomaly response action associated with the substrate processing equipment.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes high level recitation of training a machine learning model.)
Therefore, claim 9 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 9 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“and training a machine learning model using training input comprising a first window of time of the historical sensor data and target output comprising the first window of time or a second window of time of the historical sensor data to generate a trained machine learning model, the trained machine learning model to perform an anomaly response action associated with the substrate processing equipment.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes high level recitation of training a machine learning model.)
Therefore, claim 9 is subject-matter ineligible.
Regarding claim 10:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 10 recites the step:
“The method of claim 9,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 9.)
Thus, claim 10 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 10 recites the additional elements:
“wherein the target output comprises the first window of time of the historical sensor data, the target output being same as the training input.” This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 10 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 10 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“wherein the target output comprises the first window of time of the historical sensor data, the target output being same as the training input.” This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 10 is subject-matter ineligible.
Regarding claim 11:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 11 recites the step:
“The method of claim 9,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 9.)
Thus, claim 11 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 11 recites the additional elements:
“wherein the target output comprises the second window of time of the historical sensor data, the target output being offset from the training input by one or more windows of time.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 11 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 11 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“wherein the target output comprises the second window of time of the historical sensor data, the target output being offset from the training input by one or more windows of time.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes further limitation on data used for high level recitation of target output for a machine learning model.)
Therefore, claim 11 is subject-matter ineligible.
Regarding claim 12:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 12 recites the step:
“The method of claim 9,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 9.)
Thus, claim 12 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 12 recites the additional elements:
“wherein the machine learning model is a long short-term memory (LSTM) recurrent neural network (RNN) model.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes limitation refers to high level recitation of training a machine learning model.)
Therefore, claim 12 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 12 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“wherein the machine learning model is a long short-term memory (LSTM) recurrent neural network (RNN) model.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes limitation refers to high level recitation of training a machine learning model.)
Therefore, claim 12 is subject-matter ineligible.
Regarding claim 13:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 13 recites the step:
“The method of claim 12,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 12.)
“wherein the encoder determines a compressed representation of the training input,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually determine a compressed representation of the training input.)
“and wherein the decoder uses the compressed representation to predict the target output.” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually predict a target output.)
Thus, claim 13 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 13 recites the additional elements:
“wherein the LSTM RNN model comprises an encoder and a decoder,” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes limitation refers to high level recitation of training a machine learning model.)
Therefore, claim 13 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 13 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“wherein the LSTM RNN model comprises an encoder and a decoder,” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes limitation refers to high level recitation of training a machine learning model.)
Therefore, claim 13 is subject-matter ineligible.
Regarding claim 14:
Subject Matter Eligibility Analysis Step 2A Prong 1:
Claim 14 recites the step:
“The method of claim 9 further comprising:” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – refers to the mental process continued from claim 9.)
“and time windowing the trace data to generate a plurality of sequenced data sets, wherein each of the plurality of sequenced data sets corresponds to a respective time window,” (mental process - This is an observation, evaluation, judgement, or opinion, i.e. a concept performed in the human mind. See MPEP 2106.04(a)(2), III. – a user can manually determine a plurality of sequenced data sets with respect to time windows and write out the output using a pen and paper.)
Thus, claim 14 recites an abstract idea.
Subject Matter Eligibility Analysis Step 2A Prong 2:
Claim 14 recites the additional elements:
“receiving, from a plurality of sensors, trace data corresponding to the normal runs;” (This element does not integrate the abstract idea into a practical application because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g))
“wherein the training input and the target output are based on at least a subset of the plurality of sequenced data sets.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes limitation refers to data used for high level recitation of training a machine learning model.)
Therefore, claim 14 is not integrated into a practical application.
Subject Matter Eligibility Analysis Step 2B:
The additional elements of claim 14 do not provide significantly more than the abstract idea itself, taken alone and in combination because:
“receiving, from a plurality of sensors, trace data corresponding to the normal runs;” (This element does not provide significantly more because it amounts to insignificant extra-solution activity to the judicial exception (see MPEP 2106.05(g) in the form of receiving or transmitting data over a network, which is well-known, routine, and conventional (see MPEP 2106.05(d)(II)).)
“wherein the training input and the target output are based on at least a subset of the plurality of sequenced data sets.” (This element does not integrate the abstract idea into a practical application because it amounts to mere “apply it on a computer” (see MPEP 2106.05(f) Examiner notes limitation refers to data used for high level recitation of training a machine learning model.)
Therefore, claim 14 is subject-matter ineligible.
Claim 15 recites the same limitations as method claim 1, in the form of a non-transitory computer readable storage medium having instructions stored thereon, which, when executed by a processing device, cause the processing device to perform the method of claim 1. Thus, claim 15 is also directed towards performing a mental process without significantly more, therefore it is rejected under the same rationale.
Claim 16 recites the same limitations as method claim 2, in the form of a non-transitory computer readable storage medium having instructions stored thereon, which, when executed by a processing device, cause the processing device to perform the method of claim 2. Thus, claim 16 is also directed to performing a mental process without significantly more, therefore it is rejected under the same rationale.
Claim 17 recites the same limitations as method claim 3, in the form of a non-transitory computer readable storage medium having instructions stored thereon, which, when executed by a processing device, cause the processing device to perform the method of claim 3. Thus, claim 17 is also directed to performing a mental process without significantly more, therefore it is rejected under the same rationale.
Claim 18 recites the same limitations as method claim 4, in the form of a non-transitory computer readable storage medium having instructions stored thereon, which, when executed by a processing device, cause the processing device to perform the method of claim 4. Thus, claim 18 is also directed to performing a mental process without significantly more, therefore it is rejected under the same rationale.
Claim 19 recites the same limitations as method claim 5, in the form of a non-transitory computer readable storage medium having instructions stored thereon, which, when executed by a processing device, cause the processing device to perform the method of claim 5. Thus, claim 19 is also directed to performing a mental process without significantly more, therefore it is rejected under the same rationale.
Claim 20 recites the same limitations as method claim 6, in the form of a non-transitory computer readable storage medium having instructions stored thereon, which, when executed by a processing device, cause the processing device to perform the method of claim 6. Thus, claim 20 is also directed to performing a mental process without significantly more, therefore it is rejected under the same rationale.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-3, 5, 9-12, 14, 15-17, and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kocberber et al. (US 2020/0104200 A1) (hereafter referred to as Kocberber) in view of Jung et al. (US 2019/0286983 A1) (hereafter referred to as Jung).
Regarding claim 1:
Kocberber teaches:
“providing the current sensor data as input to a trained machine learning model,” (Kocberber ¶71 “After the training of the RNN LSTM is completed, the Preprocessing Module 503b will receive the raw disk drive sensor data 508 to be analyzed by the trained RNN LSTM model 507 for making predictions about disk failures.”)
“the trained machine learning model being trained using training input comprising a first window of time of historical sensor data” (Kocberber ¶50, “FIG. 3 depicts the relationship between the failure_time parameter, the heads_up_time parameter, and the last_valid_sample time parameter defining the preprocessing that establishes the training sequence of data. Thus, for a timeline of days 310, when the heads_up_time parameter 330 is specified, then, for successfully predicting an impending failure on disk failure time 320, the last valid sample to be used for training must be last_valid_sample_time 330. Any of (including all) the disk drive sensor readings obtained on or before the last_valid_sample_time may be provided in the training phase to the machine learning model. None of the disk drive sensor readings obtained after the last_valid_sample_time may be provided in the training phase.” Examiner notes the valid sample time defines the window of time of historical sensor data.)
“and target output comprising the first window of time or a second window of time of the historical sensor data to generate the trained machine learning model,” (Kocberber ¶66 “In step 410, the RNN LSTM deep learning model is trained. The model is first tuned for training using the specified hyper-parameters, and then trained using the preprocessed sequence training data. The trained model is evaluated using the preprocessed test and validation data sets.” Examiner notes the evaluation of the evaluation of the model with test and validation sets teaches target output.)
“the historical sensor data being associated with normal runs of processing of historical substrates by the substrate processing equipment;” (Kocberber ¶60 “In each created sequence (by the preprocessing stage), there is an additional field that indicates whether the sequence belongs to a healthy or failed sample.” Examiner notes a healthy run is taught by the healthy sample which is included in the data.)
“obtaining, from the trained machine learning model, one or more outputs;” (Kocberber ¶72, “Once the RNN LSTM model has been trained, the Analysis and Prediction Module 507 will receive the preprocessed input data from disk drive sensor reading, and analyze the data using the trained RNN LSTM model and provide as output, predictions regarding impending disk failures.”)
Kocberber does not distinctly disclose:
“A method comprising: identifying current sensor data associated with processing of substrates by substrate processing equipment,”
“and causing, based on the one or more outputs, an anomaly response action associated with the substrate processing equipment.”
However, Jung teaches:
“A method comprising: identifying current sensor data associated with processing of substrates by substrate processing equipment,” (Jung ¶41 “The equipment-related data refers to data which is measured through an internal/external sensor in equipment during processes from the time when the manufacturing of the semiconductor starts until the present time, such as an equipment operating time, temperature, pressure, the number of vibrations, a ratio of a specific material, or the like. That is, the equipment-related data may be referred to as input data regarding operating/state of equipment in which various physical/electrical state values regarding equipment are arranged in time series.” Examiner notes the manufacturing of the semiconductor teaches a processing of substrates by substrate processing equipment.)
“and causing, based on the one or more outputs, an anomaly response action associated with the substrate processing equipment.” (Jung ¶21 “In addition, when embodiments of the present disclosure are applied to prediction of a semiconductor manufacturing yield, a product that is predicted as not having a good yield can be specially observed/managed, and the yield prediction function can be integrated into a platform and can be used for common use in various manufacturing processes.” Examiner notes the anomaly is taught by the detection of not having a good yield which Jung responds to by specially observing or managing the product as well as integrating the prediction function into a platform for further use.)
Before the effective filing date of the invention, it would have been obvious to one of
ordinary skill in the art to combine the method of using and training a machine learning model with the input from sensors of Kocberber with the semiconductor manufacturing data from sensors of Jung in order to make time based predictions in a semiconductor manufacturing process using a neural network (Jung ¶64).
Regarding claim 2:
Kocberber as modified in claim 1 teaches all the limitations of claim 1.
Kocberber as modified further teaches:
“The method of claim 1, wherein the target output comprises the first window of time of the historical sensor data, the target output being same as the training input.” (Kocberber ¶72 “During the training phase, the output of the Preprocessing Module 503 is fed to the RNN LSTM Training Module 505 for training the RNN LSTM deep learning model. Prior to training, the Training Module 505 tunes the RNN LSTM model using hyper-parameter specifications 504 that may be provided by a user to the System 500. The Model Evaluation Module 506 is responsible for testing and validating the RNN LSTM model and establishing the trained RNN LSTM model that will then be used for analyzing the disk drive sensor data.” Examiner notes the first window of time is before the model has been trained i.e. the training phase.)
Regarding claim 3:
Kocberber as modified teaches all the limitations of claim 1.
Kocberber as modified further teaches:
“The method of claim 1, wherein the target output comprises the second window of time of the historical sensor data, the target output being offset from the training input by one or more windows of time.” (Kocberber ¶72 “Once the RNN LSTM model has been trained, the Analysis and Prediction Module 507 will receive the preprocessed input data from disk drive sensor reading, and analyze the data using the trained RNN LSTM model and provide as output, predictions regarding impending disk failures.” Examiner notes the second window of time is the sensor data obtained after training the model.)
Regarding claim 5:
Kocberber as modified teaches all the limitations of claim 1.
Kocberber as modified further teaches:
“The method of claim 1, wherein the trained machine learning model is a trained long short-term memory (LSTM) recurrent neural network (RNN) model.” (Kocberber ¶44 “The deep learning model used in embodiments described herein is an RNN LSTM machine learning model.”)
Regarding claim 9, claim 9 recites substantially similar limitations to claim 1, and is therefore rejected under the same analysis.
Regarding claim 10, claim 10 recites substantially similar limitations to claim 2, and is therefore rejected under the same analysis.
Regarding claim 11, claim 11 recites substantially similar limitations to claim 3, and is therefore rejected under the same analysis.
Regarding claim 12, claim 12 recites substantially similar limitations to claim 5, and is therefore rejected under the same analysis.
Regarding claim 14:
Kocberber as modified teaches all the limitations of claim 9.
Kocberber as modified further teaches:
“receiving, from a plurality of sensors, trace data corresponding to the normal runs;” (Kocberber ¶71 “After the training of the RNN LSTM is completed, the Preprocessing Module 503b will receive the raw disk drive sensor data 508 to be analyzed by the trained RNN LSTM model 507 for making predictions about disk failures.” Examiner notes applicant specification defines trace data as sensor data with time stamps. Fig. 7 of Kocberber discloses the data received with a timestamp (see below).
PNG
media_image1.png
628
758
media_image1.png
Greyscale
)
“and time windowing the trace data to generate a plurality of sequenced data sets, wherein each of the plurality of sequenced data sets corresponds to a respective time window, wherein the training input and the target output are based on at least a subset of the plurality of sequenced data sets.” (Kocberber ¶72 “During the training phase, the output of the Preprocessing Module 503 is fed to the RNN LSTM Training Module 505 for training the RNN LSTM deep learning model. Prior to training, the Training Module 505 tunes the RNN LSTM model using hyper-parameter specifications 504 that may be provided by a user to the System 500. The Model Evaluation Module 506 is responsible for testing and validating the RNN LSTM model and establishing the trained RNN LSTM model that will then be used for analyzing the disk drive sensor data.” Kocberber ¶72 “Once the RNN LSTM model has been trained, the Analysis and Prediction Module 507 will receive the preprocessed input data from disk drive sensor reading, and analyze the data using the trained RNN LSTM model and provide as output, predictions regarding impending disk failures.” Examiner notes the first window of time is before the model has been trained i.e. the training phase. Examiner notes the second window of time is the sensor data obtained for use after training the model.)
Regarding claim 15, Kocberber as modified in claim 1 teaches a non-transitory computer readable storage medium having instructions stored thereon, which, when executed by a processing device, cause the processing device (Kocberber ¶132-133 “For example, FIG. 9 is a block diagram that illustrates a computer system 900 upon which an embodiment of the invention may be implemented. Computer system 900 includes a bus 902 or other communication mechanism for communicating information, and a hardware processor 904 coupled with bus 902 for processing information. Hardware processor 904 may be, for example, a general purpose microprocessor. Computer system 900 also includes a main memory 906, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 902 for storing information and instructions to be executed by processor 904. Main memory 906 also may be used for storing temporary variables or other intermediate information during execution of instructions to be executed by processor 904. Such instructions, when stored in non-transitory storage media accessible to processor 904, render computer system 900 into a special-purpose machine that is customized to perform the operations specified in the instructions.”) to perform the method of claim 1 (see rejection of claim 1) and is therefore rejected under the same analysis.
Regarding claim 16, claim 16 recites substantially similar limitations to claim 2, and is therefore rejected under the same analysis.
Regarding claim 17, claim 17 recites substantially similar limitations to claim 3, and is therefore rejected under the same analysis.
Regarding claim 19, claim 19 recites substantially similar limitations to claim 5, and is therefore rejected under the same analysis.
Claim(s) 4, 7, and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kocberber et al. (US 2020/0104200 A1) (hereafter referred to as Kocberber) in view of Jung et al. (US 2019/0286983 A1) (hereafter referred to as Jung), further in view of Verma et al. (US 2019/0312898 A1) (hereafter referred to as Verma).
Regarding claim 4:
Kocberber as modified in claim 1 teaches all the limitations of claim 1.
Kocberber as modified further teaches:
“The method of claim 1, wherein: the current sensor data comprises a plurality of sequenced data sets at a first set of windows of time;” (Kocberber ¶72 “During the training phase, the output of the Preprocessing Module 503 is fed to the RNN LSTM Training Module 505 for training the RNN LSTM deep learning model. Prior to training, the Training Module 505 tunes the RNN LSTM model using hyper-parameter specifications 504 that may be provided by a user to the System 500. The Model Evaluation Module 506 is responsible for testing and validating the RNN LSTM model and establishing the trained RNN LSTM model that will then be used for analyzing the disk drive sensor data.” Examiner notes the first window of time is before the model has been trained i.e. the training phase. Kocberber ¶43 “Embodiments described herein operate with sequences of disk sensor data. The disk sensor data contain sensor readings that are collected from a disk for a certain period of time”)
“comprising predicted sequenced data sets at a second set of windows of time;” (Kocberber ¶43 “Embodiments described herein operate with sequences of disk sensor data. The disk sensor data contain sensor readings that are collected from a disk for a certain period of time”)
“and each window of time of the second set of windows of time is offset from a corresponding window of time of the first set of windows of time by one or more windows of time.” (Kocberber ¶72 “Once the RNN LSTM model has been trained, the Analysis and Prediction Module 507 will receive the preprocessed input data from disk drive sensor reading, and analyze the data using the trained RNN LSTM model and provide as output, predictions regarding impending disk failures.” Examiner notes the second window of time is the sensor data obtained after training the model.)
Kocberber as modified does not distinctly disclose:
“the one or more outputs comprise reconstruction data”
However, Verma teaches:
“the one or more outputs comprise reconstruction data” (Verma ¶68 “Training of the overall GCRNN model can be achieved by taking a chunk of sequences (e.g., m-number of sequences) of sensor data of length L from the network operating under normal behavior. In other words, a time series of length L of sensor data from the network can be used to train the models shown, to learn the normal behavior of the computer network. End to end training of the GCRNN can be achieved using back propagation, in some embodiments.”
Verma ¶69 “According to various embodiments, the device implementing architecture 400 may detect an anomaly by checking the reconstruction error of the GCRNN against a defined threshold for a given input x.sub.t. For example, such an error can be computed from {circumflex over (x)}.sub.t, the output of the ConvLSTM model. In some embodiments, the anomaly threshold may be set manually based on input from a user interface. In other embodiments, the anomaly threshold may be set based on a percentage deviation from the ‘normal’ behavior.”)
Before the effective filing date of the invention, it would have been obvious to one of
ordinary skill in the art to combine the method of using and training a machine learning model with the input from sensors of Kocberber as modified with the reconstruction data of Verma in order to detect anomalies in sensor data which deviate from normal behavior (Verma ¶69).
Regarding claim 7:
Kocberber as modified teaches all the limitations of claim 1.
Kocberber as modified does not distinctly disclose:
“The method of claim 1, wherein the causing of the anomaly response action comprises: comparing the current sensor data to reconstruction data associated with the one or more outputs to generate model reconstruction error;”
“and identifying an anomaly responsive to determining that the model reconstruction error is greater than a threshold error.”
However, Verma teaches:
“The method of claim 1, wherein the causing of the anomaly response action comprises: comparing the current sensor data to reconstruction data associated with the one or more outputs to generate model reconstruction error;” (Verma ¶69 “According to various embodiments, the device implementing architecture 400 may detect an anomaly by checking the reconstruction error of the GCRNN against a defined threshold for a given input x.sub.t. For example, such an error can be computed from {circumflex over (x)}.sub.t, the output of the ConvLSTM model. In some embodiments, the anomaly threshold may be set manually based on input from a user interface. In other embodiments, the anomaly threshold may be set based on a percentage deviation from the ‘normal’ behavior.”)
“and identifying an anomaly responsive to determining that the model reconstruction error is greater than a threshold error.” (Verma ¶69-70 “the anomaly threshold may be set based on a percentage deviation from the ‘normal’ behavior. When the device implementing architecture 400 detects an anomaly in the computer network, it may initiate any number of mitigation actions.”)
Before the effective filing date of the invention, it would have been obvious to one of
ordinary skill in the art to combine the method of using and training a machine learning model with the input from sensors of Kocberber as modified with the reconstruction data of Verma in order to detect anomalies in sensor data which deviate from normal behavior (Verma ¶69).
Regarding claim 18, claim 18 recites substantially similar limitations to claim 4, and is therefore rejected under the same analysis.
Claim(s) 6, 13, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kocberber et al. (US 2020/0104200 A1) (hereafter referred to as Kocberber) in view of Jung et al. (US 2019/0286983 A1) (hereafter referred to as Jung), further in view of Shin (US 2019/0129947 A1) (hereafter referred to as Shin).
Regarding claim 6:
Kocberber as modified teaches all the limitations of claim 5.
Kocberber further teaches:
“wherein the input comprises a current plurality of sequenced data sets,” (Kocberber ¶43 “Embodiments described herein operate with sequences of disk sensor data. The disk sensor data contain sensor readings that are collected from a disk for a certain period of time”)
Kocberber as modified does not distinctly disclose:
“The method of claim 5, wherein the trained LSTM RNN model comprises an encoder and a decoder,”
“wherein the encoder determines a compressed representation of the input,”
“wherein the decoder uses the compressed representation to predict a future plurality of sequenced data sets.”
However, Shin teaches:
“The method of claim 5, wherein the trained LSTM RNN model comprises an encoder and a decoder,” (Shin ¶3 “The encoder-decoder mechanism-based NMT refers to an artificial neural network learning and prediction mechanism which uses a recurrent neural network with long short-term memory (RNN-LSTM) or a convolutional neural network (CNN) to compress (or abstract) an input sentence of source language into a single or multiple N-dimensional vectors by using an encoder and to generate an output sentence (a translation result) of target language from the compressed (abstracted) representations by using a decoder.”)
“wherein the encoder determines a compressed representation of the input,” (Shin ¶3 “The encoder-decoder mechanism-based NMT refers to an artificial neural network learning and prediction mechanism which uses a recurrent neural network with long short-term memory (RNN-LSTM) or a convolutional neural network (CNN) to compress (or abstract) an input sentence of source language into a single or multiple N-dimensional vectors by using an encoder and to generate an output sentence (a translation result) of target language from the compressed (abstracted) representations by using a decoder.”)
“wherein the decoder uses the compressed representation to predict a future plurality of sequenced data sets.” (Shin ¶3 “The encoder-decoder mechanism-based NMT refers to an artificial neural network learning and prediction mechanism which uses a recurrent neural network with long short-term memory (RNN-LSTM) or a convolutional neural network (CNN) to compress (or abstract) an input sentence of source language into a single or multiple N-dimensional vectors by using an encoder and to generate an output sentence (a translation result) of target language from the compressed (abstracted) representations by using a decoder.”)
Before the effective filing date of the invention, it would have been obvious to one of
ordinary skill in the art to combine the method of using and training a machine learning model of Kocberber as modified with the encoder and decoder of Shin because of the extensive research and widespread support by various companies for implementing LSTMs with an encoder-decoder based mechanism (Shin ¶3).
Regarding claim 13, claim 13 recites substantially similar limitations to claim 6, and is therefore rejected under the same analysis.
Regarding claim 20, claim 20 recites substantially similar limitations to claim 6, and is therefore rejected under the same analysis.
Claim(s) 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kocberber et al. (US 2020/0104200 A1) (hereafter referred to as Kocberber) in view of Jung et al. (US 2019/0286983 A1) (hereafter referred to as Jung), further in view of Verma et al. (US 2019/0312898 A1) (hereafter referred to as Verma) as applied to claim 7, further in view of Yadav et al. (https://arxiv.org/pdf/1605.01534v1) (hereafter referred to as Yadav).
Regarding claim 8:
Kocberber as modified teaches all the limitations of claim 7.
Kocberber as modified does not distinctly disclose:
“generating a plurality of anomaly scores based on the one or more outputs, wherein each of the plurality of anomaly scores corresponds to a respective sensor of a plurality of sensors; and ranking contribution to the model reconstruction error by each of the plurality of sensors based on the plurality of anomaly scores.”
However, Yadav teaches:
“generating a plurality of anomaly scores based on the one or more outputs, wherein each of the plurality of anomaly scores corresponds to a respective sensor of a plurality of sensors; and ranking contribution to the model reconstruction error by each of the plurality of sensors based on the plurality of anomaly scores.” (Yadav §2 ¶1, “Consider a multivariate time-series
X
=
x
1
,
x
2
,
…
,
x
n
, where each point
x
t
∈
R
m
in the time series is an m-dimensional vector of variables/sensors. Time series data arising from a controlled dynamical system can usually be divided into two sets of variables:
X
c
being a set `control' variables such as accelerator pedal position (APP), and
X
d
which represents `dependent' variables such as coolant temperature (CT), torque, etc.” Yadav §2.3 ¶1, “We learn a stacked / deep LSTM based prediction model where
x
t
is used to predict
{
x
t
+
1
,
…
,
x
t
+
l
}
, i.e., time-series for next
l
time-steps. An error vector
ⅇ
t
for point
x
t
is given by
ⅇ
t
=
e
1
t
,
e
2
t
,
…
,
e
l
t
, where
e
i
t
is the difference between
x
t
and its value as predicted at time
t
-
ⅈ
. The likelihood of a point
x
t
being normal is given by the likelihood score of the corresponding error vector
ⅇ
t
computed from a learned Gaussian probability density function over the set of error vectors from the normal data. The parameters of the Gaussian distribution are estimated using Maximum Likelihood Estimation over a set of error vectors from normal time-series data. Further, a threshold on likelihood scores is estimated by maximizing F-score so that points with likelihood score below the threshold are considered to be anomalous points. Different validation sets were used to avoid over-fitting while learning network parameters, prediction length and threshold for likelihood scores. For further details on LSTM-AD, readers can look at [4].” Examiner notes the plurality of scores calculated which are used when determining anomalies. Examiner also notes the values come from different sensors as taught in Yadav §2. Examiner further notes these scores are used for ranking as seen in Table 1 of Yadav (see below)
PNG
media_image2.png
164
300
media_image2.png
Greyscale
)
Before the effective filing date of the invention, it would have been obvious to one of
ordinary skill in the art to combine the method of using and training a machine learning model of Kocberber as modified with the method using plurality of scores and sensors of Yadav in order to overcome limitations caused by limited training data (Yadav §1 ¶3).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Malhotra et al. (https://arxiv.org/pdf/1607.00148) also discloses a method of using an LSTM model for anomaly detection from sensor data.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Peter T Annis whose telephone number is (571)270-1059. The examiner can normally be reached M-F, 7:30am to 5pm ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Alexey Shmatov can be reached at (571) 270-3428. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/PETER THOMAS ANNIS/Examiner, Art Unit 2123
/ALEXEY SHMATOV/Supervisory Patent Examiner, Art Unit 2123
Prosecution Insights