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 .
Priority
Application is a continuation of Provisional Application No. 63/127,832, filed
on December 18, 2020.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 06/15/2023is in compliance
with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being
considered by the examiner.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-3, 7-11, and 16-18 are rejected under 35 U.S.C. 103 as being unpatentable over Jiang et al. “Classification for Incomplete Data Using Classifier Ensembles” 10/15/2005 in view of Yati Pub. No.: US 20190004504 A1.
Regarding Claim 1 Jiang teaches A non-transitory computer-readable medium having instructions that, when executed by a computer, cause the computer to execute a method for training a defect location prediction model, the method comprising: 1…wherein the set of locations comprise locations with partial datasets in which data regarding one or more of the process-related parameters is absent2… (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters))
processing the datasets to generate multiple parameter groups having different sets of process- related parameters, wherein each parameter group includes data for each parameter of a corresponding set of process-related parameters; (Jiang, page 560-561, section III, teaches the splitting of the data into groups of data that is used in the training of the sub models of the ensemble learning model. The data that is contained in each set of data is sampled from an input space that can represent a location or a parameter group.)
and for each parameter group: creating a sub-model 2… based on the corresponding set of process-related parameters of the parameter group; and training the sub-model by using data from the parameter group. (Jiang, page 560-561, section III, teaches the use of the groups of data to train sub-models that are used in the ensemble classification model (i.e. a prediction model))
Jiang does not teach 1…receiving a dataset for each of a set of locations on a set of substrates having data regarding a plurality of process-related parameters,…; However, Yati in analogous art teaches this limitation (Yati, paragraph 0031-0035, teaches a semiconductor defect detection system that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064]))
Further Jiang does not teach 2…of the defect location prediction model… However, Yati in analogous art teaches this limitation (Yati, paragraph 0031-0035, teaches a semiconductor defect detection system that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064]))
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to combine Yati’s teaching of training a prediction model to predict a substrate defect location with Jiang’s teaching of an ensemble training method to train an ensemble on partial data. The motivation to do so would be that it is common in the art to combine multiple models in an ensemble to create more accurate models and standard model struggle with processing missing data which ensembles help midigate (Jiang, page 559, section I).
Regarding claim 2 The combination of Jiang and Yati teaches the computer-readable medium of claim 1, wherein training the sub-model is an iterative process in which each iteration includes: inputting data from the parameter group to the sub-model to obtain a predicted result from the sub-model, (Jiang, page 560-561, section III, teaches the inputting of groups of data into the sub models of the ensemble learning model to obtain a prediction from the sub model.)
wherein the predicted result of the sub-model is indicative of whether a specified location on a specified substrate is likely to be defective or non-defective; (Yati, paragraph 0011, 0034-0036, teaches the inputting of a data set to the defect prediction model and receiving a prediction as to whether or not a specified substrate is likely to have a defect or not.)
determining a cost function based on the predicted result and an actual result that is provided as input associated with the parameter group; and adjusting the sub-model based on the cost function. (Yati, paragraph 0011, 0036-0040, teaches the predicting of defect locations based on training data that has associated validated defect result, the system uses tensor flow for its deep learning model which uses a loss function that compares a y_true value (i.e. actual result) and a y_pred value (i.e. predicted result) these values are compared and used in the loss function to adjust the model throughout the training process. (https://web.archive.org/web/20200322091733/https://www.tensorflow.org/api_docs/python/tf/keras/losses/Loss))
Regarding claim 3 The combination of Jiang and Yati teaches The computer-readable medium of claim 2, wherein the actual result is an inspection result of the specified substrate obtained from an inspection system, the actual result indicative of whether the specified location is defective or non-defective. (Yati, paragraph 0036, teaches the use of a review tool to check the predicted locations for defects to validate the predicted result from the machine learning model to check its accuracy.)
Regarding Claim 7 The combination of Jiang and Yati teaches The computer-readable medium of claim 1, wherein processing the datasets includes: selecting a first set of process-related parameters from the plurality of process-related parameters to generate a first parameter group; (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters) Further Jiang, page 560-561, section III, teaches the splitting of the data into groups of data that is used in the training of the sub models of the ensemble learning model. The data that is contained in each set of data is sampled from an input space that can represent a location or a parameter group.)
and populating the first parameter group with data for the first set of process-related parameters from the datasets, wherein the datasets that do not have data for the first set of process-related parameters are excluded. (Jiang, page 360-362, section III, teaches the splitting of the data sets based on the features (i.e. parameters) pf the data set and allows for the splitting to be done in that data that isn’t related to the same feature is excluded from the dataset.)
Regarding Claim 8 The combination of Jiang and Yati teaches The computer-readable medium of claim 1, wherein training the sub-models includes: training a first sub-model corresponding to a first parameter group by inputting data from the first parameter group, the first parameter group including data for a first set of process-related parameters from the datasets; and training a second sub-model corresponding to a second parameter group using the first sub-model, wherein the second parameter group includes one or more parameters in addition to the first set of process-related parameters (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters) Further Jiang, page 560-561, section III, teaches the splitting of the data into groups of data that is used in the training of the sub models of the ensemble learning model. The data that is contained in each set of data is sampled from an input space that can represent a location or a parameter group. The data in these groups can be groups of data from another group with additional data added.)
Regarding Claim 9 The combination of Jiang and Yati teaches The computer-readable medium of claim 1, wherein each sub-model includes two or more process-related parameters. (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters) Further Jiang, page 560-561, section III, teaches the splitting of the data into groups of data that is used in the training of the sub models of the ensemble learning model. The data that is contained in each set of data is sampled from an input space that can represent a location or a parameter group. The data that the sub models are trained on can include two or more features (i.e. process-related parameters).
Regarding Claim 10 The combination of Jiang and Yati teaches The computer-readable medium of claim 1, wherein the process-related parameters include parameters associated with multiple processes involved in forming a pattern on a substrate. (Yati, paragraph 0043, teaches the parameters that are input into the deep learning model to be that of critical dimensions (i.e. process parameters), including one of a design, a care area, or a design clip; tool parameters (e.g., focus, exposure))
Regarding Claim 11 The combination of Jiang and Yati teaches The computer-readable medium of claim 10, wherein the parameters include metrology data associated with the multiple processes. (Yati, paragraph 0043, teaches the parameters that are input into the deep learning model to be that of critical dimensions (i.e. process parameters), including one of a design, a care area, or a design clip; tool parameters (e.g., focus, exposure))
Regarding Claim 16 The combination of Jiang teaches An apparatus for training a defect location prediction model to predict a defect on a substrate, the apparatus comprising: a memory storing a set of instructions; and at least one processor configured to execute the set of instructions to cause the apparatus to perform operations comprising: 1…wherein the set of locations comprise locations with partial datasets in which data regarding one or more of the process-related parameters is absent; (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters))
processing the datasets to generate multiple parameter groups having different sets of process-related parameters, wherein each parameter group includes data for each parameter of a corresponding set of process-related parameters; (Jiang, page 560-561, section III, teaches the splitting of the data into groups of data that is used in the training of the sub models of the ensemble learning model. The data that is contained in each set of data is sampled from an input space that can represent a location or a parameter group.)
and for each parameter group: creating a sub-model 2… based on the corresponding set of process-related parameters of the parameter group; and training the sub-model by using data from the parameter group. (Jiang, page 560-561, section III, teaches the use of the groups of data to train sub-models that are used in the ensemble classification model (i.e. a prediction model))
Jiang does not teach 1… receiving a dataset for each of a set of locations on a set of substrates having data regarding a plurality of process-related parameters, … However, Yati in analogous art teaches this limitation (Yati, paragraph 0031-0035, teaches a semiconductor defect detection system that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064]))
Further Jiang does not teach 2… of the defect location prediction model… However, Yati in analogous art teaches this limitation (Yati, paragraph 0031-0035, teaches a semiconductor defect detection system that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064]))
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to combine Yati’s teaching of training a prediction model to predict a substrate defect location with Jiang’s teaching of an ensemble training method to train an ensemble on partial data. The motivation to do so would be that it is common in the art to combine multiple models in an ensemble to create more accurate models and standard model struggle with processing missing data which ensembles help midigate (Jiang, page 559, section I).
Regarding Claim 17 The combination of Jiang and Yati teaches The apparatus of claim 16, wherein training the sub-model is an iterative process in which each iteration includes: inputting data from the parameter group to the sub-model to obtain a predicted result from the sub-model, (Jiang, page 560-561, section III, teaches the inputting of groups of data into the sub models of the ensemble learning model to obtain a prediction from the sub model.)wherein the predicted result of the sub-model is indicative of whether a specified location on a specified substrate is likely to be defective or non-defective; (Yati, paragraph 0011, 0034-0036, teaches the inputting of a data set to the defect prediction model and receiving a prediction as to whether or not a specified substrate is likely to have a defect or not.) determining a cost function based on the predicted result and an actual result that is provided as input associated with the parameter group; and adjusting the sub-model based on the cost function. (Yati, paragraph 0011, 0036-0040, teaches the predicting of defect locations based on training data that has associated validated defect result, the system uses tensor flow for its deep learning model which uses a loss function that compares a y_true value (i.e. actual result) and a y_pred value (i.e. predicted result) these values are compared and used in the loss function to adjust the model throughout the training process. (https://web.archive.org/web/20200322091733/https://www.tensorflow.org/api_docs/python/tf/keras/losses/Loss))
Regarding Claim 18 The combination of Jiang and Yati teaches he apparatus of claim 17, wherein the actual result is an inspection result of the specified substrate obtained from an inspection system, the actual result indicative of whether the specified location is defective or non-defective. (Yati, paragraph 0036, teaches the use of a review tool to check the predicted locations for defects to validate the predicted result from the machine learning model to check its accuracy.)
Claims 4-6 and 19-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Jiang et al. “Classification for Incomplete Data Using Classifier Ensembles” 10/15/2005 in view of Yati Pub. No.: US 20190004504 A1 in further view of Garcia et al. “Dynamic ensemble selection for multi-class imbalanced datasets” June 2018.
Regarding Claim 4 The combination of Jiang and Yati teaches The computer-readable medium of claim 1 further comprising: receiving a first partial dataset for a first location on a first substrate; (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters). The models are then used to make classifications based on newly received partial datasets.) 1… to predict a defect for the first location based on the first partial dataset (Yati, paragraph 0031-0035, teaches a semiconductor defect detection model that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064])).
The combination of Jiang and Yati does not teach 1…selecting one of the sub-models based on a first set of process-related parameters available in the first partial dataset; However, Garcia in analogous art teaches this limitation (Garcia, page 25 section 3, and pages 27-29, section 4.2.2, teaches the use of a dynamic ensemble selection model in which the ensemble selects the best subset of models for the current input dataset. This dynamic ensemble selection can select a subset of models in the ensemble, where the subset can be a subset containing a single model to perform the current prediction task.)
Further The combination of Jiang and Yati does not teach and executing the selected sub-model…. However, Garcia in analogous art teaches this limitation (Garcia, pages 27-29, section 4.2.2, and pages 29-32, section 5, teach the executing of the selected subset of models, which can be a subset containing a single model, to perform a prediction task.)
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to combine Garcia’s teaching of Dynamically selecting subsets in an ensemble model with the combination of Jiang and Yati’s teaching of an ensemble model for predicting defects based on partial datasets. The motivation to do so would be to to help improve the prediction of the ensemble model by selecting the best sub models that most accurately represent the same attributes as the input data. (Garcia, page 22-23, Section I)
Regarding Claim 5 The combination of Jiang and Yati teaches The computer-readable medium of claim 1 further comprising: receiving a first partial dataset for a first location on a first substrate, wherein the first partial dataset includes data for a first set of process-related parameters of the plurality of process-related parameters; (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters). The models are then used to make classifications based on newly received partial data) 1…generate a prediction of a defect for the first location by inputting a portion of the first partial dataset corresponding to parameters of the sub-model; 2…a defect for the first location… (Yati, paragraph 0031-0035, teaches a semiconductor defect detection model that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064])).
The combination of Jiang and Yati teaches 1…selecting a set of sub-models, wherein each sub-model of the set corresponds to different parameter subsets of the first set of process-related parameters; However, Garcia in analogous art teaches this limitation (Garcia, page 25 section 3, and pages 27-29, section 4.2.2, teaches the use of a dynamic ensemble selection model in which the ensemble selects the best subset of models for the current input dataset. This dynamic ensemble selection can select a subset of models in the ensemble, where the subsets are selected based on the parameters that are present in the input data.)
Further The combination of Jiang and Yati teaches for each sub-model of the set, executing the sub-model to … and executing an ensemble model to predict 2…. based on the predictions generated by the set of sub-models However, Garcia in analogous art teaches this limitation (Garcia, pages 27-29, section 4.2.2, and pages 29-32, section 5, teach the executing of the selected subset of models, to perform a prediction task and then aggregating the results of all the sub-models to get the final prediction from the ensemble of sub-models.)
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to combine Garcia’s teaching of Dynamically selecting subsets in an ensemble model with the combination of Jiang and Yati’s teaching of an ensemble model for predicting defects based on partial datasets. The motivation to do so would be to to help improve the prediction of the ensemble model by selecting the best sub models that most accurately represent the same attributes as the input data. (Garcia, page 22-23, Section I)
Regarding Claim 6 The combination of Jiang, Yati, and Garcia teaches The computer-readable medium of claim 5, wherein the ensemble model (Jiang, page 560-561, section III, teaches the inputting of groups of data into the sub models of the ensemble learning model to obtain a prediction from the sub model.)
is trained to predict a defect for a location on a substrate based on an initial dataset (Yati, paragraph 0031-0035, teaches a semiconductor defect detection model that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064])).
That includes predictions generated by the set of sub-models (Jiang, page 561, section III-B, teaches the training of a ensemble classifier that is trained on predictions that are generated by the sub models of the ensemble model.)
for a number of locations on a number of substrates. (Yati, paragraph 0031-0035, teaches a semiconductor defect detection model that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064])).
Regarding Claim 19 The combination of Jiang and Yati teaches The apparatus of claim 16, wherein the operations further comprise: receiving a first partial dataset for a first location on a first substrate; (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters). The models are then used to make classifications based on newly received partial datasets.) 1… to predict a defect for the first location based on the first partial dataset. (Garcia, page 25 section 3, and pages 27-29, section 4.2.2, teaches the use of a dynamic ensemble selection model in which the ensemble selects the best subset of models for the current input dataset. This dynamic ensemble selection can select a subset of models in the ensemble, where the subset can be a subset containing a single model to perform the current prediction task.)
The combination of Jiang and Yati does not teach 1…selecting one of the sub-models based on a first set of process-related parameters available in the first partial dataset; However, Garcia in analogous art teaches this limitation (Garcia, page 25 section 3, and pages 27-29, section 4.2.2, teaches the use of a dynamic ensemble selection model in which the ensemble selects the best subset of models for the current input dataset. This dynamic ensemble selection can select a subset of models in the ensemble, where the subset can be a subset containing a single model to perform the current prediction task.)
Further The combination of Jiang and Yati does not teach and executing the selected sub-model… However, Garcia in analogous art teaches this limitation (Garcia, pages 27-29, section 4.2.2, and pages 29-32, section 5, teach the executing of the selected subset of models, which can be a subset containing a single model, to perform a prediction task.)
Regarding Claim 20 The combination of Jiang and Yati teaches The apparatus of claim 16, wherein the operations further comprise: receiving a first partial dataset for a first location on a first substrate, wherein the first partial dataset includes data for a first set of process-related parameters of the plurality of process-related parameters; (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters). The models are then used to make classifications based on newly received partial data) 1… generate a prediction of a defect for the first location by inputting a portion of the first partial dataset corresponding to parameters of the sub-model; 2…to predict a defect for the first location …(Yati, paragraph 0031-0035, teaches a semiconductor defect detection model that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064])).
The combination of Jiang and Yati does not teach 1… selecting a set of sub-models, wherein each sub-model of the set corresponds to different parameter subsets of the first set of process-related parameters; However, Garcia in analogous art teaches this limitation (Garcia, page 25 section 3, and pages 27-29, section 4.2.2, teaches the use of a dynamic ensemble selection model in which the ensemble selects the best subset of models for the current input dataset. This dynamic ensemble selection can select a subset of models in the ensemble, where the subsets are selected based on the parameters that are present in the input data.)
Further The combination of Jiang and Yati does not teach For each sub-model of the set, 2… and executing an ensemble model …based on the predictions generated by the set of sub-models. However, Garcia in analogous art teaches this limitation (Garcia, pages 27-29, section 4.2.2, and pages 29-32, section 5, teach the executing of the selected subset of models, to perform a prediction task and then aggregating the results of all the sub-models to get the final prediction from the ensemble of sub-models.)
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to combine Garcia’s teaching of Dynamically selecting subsets in an ensemble model with the combination of Jiang and Yati’s teaching of an ensemble model for predicting defects based on partial datasets. The motivation to do so would be to to help improve the prediction of the ensemble model by selecting the best sub models that most accurately represent the same attributes as the input data. (Garcia, page 22-23, Section I)
Claims 12-15 are rejected under 35 U.S.C. 103 as being unpatentable over Jiang et al. “Classification for Incomplete Data Using Classifier Ensembles” 10/15/2005 in view of Yati Pub. No.: US 20190004504 A1 and Garcia et al. “Dynamic ensemble selection for multi-class imbalanced datasets” June 2018.
Regarding Claim 12 Jiang teaches A non-transitory computer-readable medium having instructions that, when executed by a computer, cause the computer to execute a method for predicting a defect at a location on a substrate, the method comprising: receiving a partial dataset for1…, wherein the partial dataset includes data for a subset of a set of process-related parameters; 2… (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters) Further Jiang, page 560-561, section III, teaches the splitting of the data into groups of data that is used in the training of the sub models of the ensemble learning model. The data that is contained in each set of data is sampled from an input space that can represent a location or a parameter group.)
Jiang does not teach 2…selecting a first sub-model from a plurality of sub-models of 3…, wherein the first sub-model is selected based on process-related parameters available in the partial dataset; However, Garcia in analogous art teaches this limitation (Garcia, page 25 section 3, and pages 27-29, section 4.2.2, teaches the use of a dynamic ensemble selection model in which the ensemble selects the best subset of models for the current input dataset. This dynamic ensemble selection can select a subset of models in the ensemble, where the subset can be a subset containing a single model to perform the current prediction task.) and executing the selected sub-model 4… (Garcia, pages 27-29, section 4.2.2, and pages 29-32, section 5, teach the executing of the selected subset of models, which can be a subset containing a single model, to perform a prediction task.)
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to combine Garcia’s teaching of Dynamically selecting subsets in an ensemble model with the combination of Jiang teaching of an ensemble model for predicting defects based on partial datasets. The motivation to do so would be to help improve the prediction of the ensemble model by selecting the best sub models that most accurately represent the same attributes as the input data. (Garcia, page 22-23, Section I)
Further the combination of Jiang and Garcia does not teach 1… a location on a substrate… 3… a defect location prediction model trained to predict a defect associated with the location on the substrate… 4… to predict the defect. … However, Yati in analogous art teaches this limitation (Yati, paragraph 0031-0035, teaches a semiconductor defect detection system that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064]))
It would have been obvious to one skilled in the art before the effective filing date of the claimed invention to combine Yati’s teaching of training a prediction model to predict a substrate defect location with Jiang and Garcia’s teaching of an ensemble training method to train an ensemble on partial data. The motivation to do so would be that it is common in the art to combine multiple models in an ensemble to create more accurate models and standard model struggle with processing missing data which ensembles help midigate (Jiang, page 559, section I).
Regarding Claim 13 The combination of Jiang, Yati, and Garcia teaches The computer-readable medium of claim 12, wherein selecting the first sub- model includes: selecting one of the sub-models associated with a set of process-related parameters matching the process-related parameters available in the partial dataset as the first sub-model. (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters) Further Jiang, page 560-561, section III, teaches the splitting of the data into groups of data that is used in the training of the sub models of the ensemble learning model. The data that is contained in each set of data is sampled from an input space that can represent a location or a parameter group.)
Regarding Claim 14 The combination of Jiang, Yati, and Garcia teaches The computer-readable medium of claim 12, wherein selecting the first sub- model further includes: selecting a set of sub-models, (Garcia, page 25 section 3, and pages 27-29, section 4.2.2, teaches the use of a dynamic ensemble selection model in which the ensemble selects the best subset of models for the current input dataset. This dynamic ensemble selection can select a subset of models in the ensemble, where the subsets are selected based on the parameters that are present in the input data.)
wherein each sub-model of the set corresponds to different process-related parameters available in the partial dataset; for each sub-model of the set, (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters). The models are then used to make classifications based on newly received partial data) executing the corresponding sub-model to generate a prediction of a defect for the location (Yati, paragraph 0031-0035, teaches a semiconductor defect detection model that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064])). by inputting a portion of the partial dataset corresponding to process-related parameters of the sub-model; (Jiang, page 560-561, section III, teaches the teaches the training of ensemble learning model that receives partial datasets uses them to train classification models. The data is gathered from different sample spaces (i.e. locations) and each space contains features (i.e. process-related parameters). The models are then used to make classifications based on newly received partial data and executing an ensemble model (Garcia, pages 27-29, section 4.2.2, and pages 29-32, section 5, teach the executing of the selected subset of models, to perform a prediction task and then aggregating the results of all the sub-models to get the final prediction from the ensemble of sub-models.) to predict a defect for the location based on the predictions generated by the set of sub-models. (Yati, paragraph 0031-0035, teaches a semiconductor defect detection model that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064])).
Regarding Claim 15 The combination of Jiang, Yati, and Garcia teaches The computer-readable medium of claim 14, wherein the ensemble model (Jiang, page 560-561, section III, teaches the inputting of groups of data into the sub models of the ensemble learning model to obtain a prediction from the sub model.)
is trained to predict a defect for a specified location on a specified substrate based on an initial dataset (Yati, paragraph 0031-0035, teaches a semiconductor defect detection model that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064])). that includes predictions generated by the sub-models (Jiang, page 561, section III-B, teaches the training of a ensemble classifier that is trained on predictions that are generated by the sub models of the ensemble model.) for a number of locations on a number of substrates. (Yati, paragraph 0031-0035, teaches a semiconductor defect detection model that receives a set of data for a plurality of defect locations, this data includes the critical dimensions of the semiconductors (i.e. process-related parameters), where the location data is from a substrate (paragraph [0064] – [0064])).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to THOMAS B LANE whose telephone number is (571)272-1872. The examiner can normally be reached M-Th: 7am-5pm; F: Out of Office.
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, MARIELA REYES can be reached at (571) 270-1006. 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.
/THOMAS BERNARD LANE/ Examiner, Art Unit 2142
/HAIMEI JIANG/ Primary Examiner, Art Unit 2142
Prosecution Insights