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 .
Information Disclosure Statement
The information disclosure statements (IDS) submitted on March 3, 2024, is being considered by the examiner.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1,4,5,9,12,13,21 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Burton et al. (US 20230071467 A1, hereinafter Burton).
Regarding claim 1, Burton teaches a method for determining confidence (Metric; See paragraph 10) for a system having two or more cascaded models (Two or more models; Paragraph 33), the method comprising: receiving a first validation data set (First Biophysical Signal Data Set; See Fig. 3a; paragraph 45) for validating performance of an upstream model
(First of the Two or more models; Paragraph 33) of the two or more cascaded models (See Paragraph 33; “… two or more models…” ) and receiving a second validation data set (Second Biophysical Data Set; See Fig. 3a; paragraph 45) for validating performance of a downstream model (Second of the Two or more models; Paragraph 33) of the two or more cascaded models wherein the second validation data set is different than the first validation set ( The Biophysical data sets contain different sets of health data recorded at the same time, as they are looking for different probabilities of different diseases.);
Burton does not explicitly teach but does anticipate estimating one or more system-level errors (Values; See Fig.3a; Paragraph 33) caused by predictions (Plurality of features or Parameters; See Fig 3a) of the upstream model based on the first validation data set (First Biophysical Signal data Set); (Burton anticipates this because they average the first values and model with the second values and model, meaning the system-level errors (Values) had to be calculated separately then combined for the average; Paragraph 33);
Burton does not explicitly teach but does anticipate estimating one or more system-level errors (Values) caused by predictions (Plurality of features or Parameters; See Fig 3a) of the downstream model based on the second validation data set (Second Biophysical Signal data set); (Burton anticipates this because they average the first values and model with the second values and model, meaning the system-level errors (Values) had to be calculated separately then combined for the average; Paragraph 33);
Burton also teaches generating a confidence interval (Averaged model; See Paragraph 33; “wherein the two or more models are combined (e.g., averaged) in an ensemble model that outputs the estimated value for the presence of abnormal left-ventricular end diastolic
pressure (LVEDP)”) that indicates a confidence (Value or Values; See Paragraph 33; “estimated value for the presence of abnormal left-ventricular end diastolic pressure (LVEDP)”; In the Specification they mention using system errors (Values) to detect illness and disease which is what Burton teaches.) for the system based on the one or more system-level errors (values; See Fig.3a; Paragraph 33) caused by predictions (Plurality of features or Parameters; See Fig 3a) of the upstream model (First model) and based on the one or more system-level errors (values) caused by predictions of the (Second Model) downstream model.
Regarding Claim 4, Burton teaches a method for determining confidence (Metric) for a system having two or more cascaded models according to claim 1 (and thus the rejection of claim 1 is incorporated).
Burton further teaches wherein generating the confidence interval (Averaged Model) further comprises: estimating one or more empirical quantiles (“estimate one or more values associated with a presence of an expected disease state or condition”; See paragraph 45) at a predefined probability level (The predefined probability level in Burtons case is the avg. probability level of a chosen disease.).
Regarding Claim 5, most of the limitations of this claim have been noted in the rejection of claim 1 and claim 4 (and thus the rejections of claim 1 and claim 4 are incorporated).
Burton further teaches wherein generating the confidence interval (Averaged model) further comprises: determining empirical error distribution (outlier detection analysis; See paragraph 45; Outlier detection analysis would detect error and error rate in a given dataset.)
for the first validation data set (First Biophysical signal Data Set) and the second validation data set (Second biophysical signal Data set).
Regarding claim 9, Burton teaches a computing system for determining confidence (Metric) for a system having two or more cascaded models (Two or more models), the computing system comprising: processing circuitry in communication with storage media, the processing circuitry configured to execute a machine learning system (See paragraph
67; “comprising one or more processors; and one or more memory having instructions respectively stored thereon, wherein execution of the instructions by the one or more processors cause the one or more processors to perform any one of the above -discussed method”) configured to: Receive a first validation data set (First Biophysical Signal Data Set; See Fig. 3a; paragraph 45) for validating performance of an upstream model (First of the Two or more models; Paragraph 33) of the two or more cascaded models (See Paragraph 33; “… two or more models…” ) and receiving a second validation data set (Second Biophysical Data Set; See Fig. 3a; paragraph 45) for validating performance of a downstream model (Second of the Two or more models; Paragraph 33) of the two or more cascaded models wherein the second validation data set is different than the first validation set (They are labeled as different Biophysical Signal Data sets, so they are different.);
Burton does not explicitly teach but does anticipate estimating one or more system-
level errors (values; See Fig.3a; Paragraph 33) caused by predictions (Plurality of features or Parameters; See Fig 3a) of the upstream model based on the first validation data set (First Biophysical Signal Data Set); (Burton anticipates this because they average the first values and
model with the second values and model, meaning the system-level errors (value) had to be calculated separately then combined for the average; Paragraph 33);
Burton does not explicitly teach but does anticipate estimating one or more system-level errors (values) caused by predictions (Plurality of features or Parameters; See Fig 3a) of the downstream model based on the second validation data set (Second Biophysical Signal data set); (Burton anticipates this because they average the first values and model with the second values and model, meaning the system-level errors (value) had to be calculated separately then combined for the average; Paragraph 33);
Burton also teaches generating a confidence interval (Averaged model; See Paragraph 33; “wherein the two or more models are combined (e.g., averaged) in an ensemble model that outputs the estimated value for the presence of abnormal left-ventricular end diastolic pressure (LVEDP)”) that indicates a confidence (Metric; See paragraph 10) for the system based on the one or more system-level errors (values; See Fig.3a; Paragraph 33) caused by predictions (Plurality of features or Parameters; See Fig 3a) of the upstream model (First model) and based on the one or more system-level errors (values) caused by predictions of the (Second Model) downstream model.
Regarding Claim 12, Burton teaches a computer system for determining confidence (Metric) for a system having two or more cascaded models according to claim 9 (and thus the rejection of claim 9 is incorporated).
Burton further teaches wherein generating the confidence interval (Averaged Model)
further comprises: estimating one or more empirical quantiles (“estimate one or more values
associated with a presence of an expected disease state or condition”; See paragraph 45) at a predefined probability level (The predefined probability level in Burtons case is the avg. probability level of disease.).
Regarding Claim 13, most of the limitations of this claim have been noted in the rejection of claim 9 and claim 12 (and thus the rejections of claim 9 and claim 12 are incorporated).
Burton further teaches wherein generating the confidence interval (Averaged model) further comprises: determining empirical error distribution (outlier detection analysis; See paragraph 45; Outlier detection analysis would detect error and error rate in a given dataset.) for the first validation data set (First Biophysical signal Data Set) and the second validation data set (Second biophysical signal Data set).
Regarding claim 21, Burton as applied to claim 9 teaches the claimed computer readable medium functionality.
Burton further teaches receiving a first validation data set (First Biophysical Signal Data Set; See Fig. 3a; paragraph 45) for validating performance of an upstream model (First of the Two or more models; Paragraph 33) of the two or more cascaded models ( See Paragraph 33; “… two or more models…” ) and receiving a second validation data set (Second Biophysical Data Set; See Fig. 3a; paragraph 45) for validating performance of a downstream model (Second of the Two or more models; Paragraph 33) of the two or more cascaded models wherein the second validation data set is different than the first validation set (They are labeled as different Biophysical Signal Data sets, so they are different.);
Burton does not explicitly teach but does anticipate estimating one or more system-level errors (values; See Fig.3a; Paragraph 33) caused by predictions (Plurality of features or Parameters; See Fig 3a) of the upstream model based on the first validation data set (First Biophysical Signal data Set); (Burton anticipates this because they average the first values and model with the second values and model, meaning the system-level errors (value) had to be calculated separately then combined for the average; Paragraph 33);
Burton does not explicitly teach but does anticipate estimating one or more system-level errors (values) caused by predictions (Plurality of features or Parameters; See Fig 3a) of the downstream model based on the second validation data set (Second Biophysical Signal data set); (Burton anticipates this because they average the first values and model with the second values and model, meaning the system-level errors (values) had to be calculated separately then combined for the average; Paragraph 33);
Burton also teaches generating a confidence interval (Averaged model; See Paragraph 33; “wherein the two or more models are combined (e.g., averaged) in an ensemble model that outputs the estimated value for the presence of abnormal left-ventricular end diastolic pressure (LVEDP)”) that indicates a confidence (Metric) for the system based on the one or more system-level errors (values; See Fig.3a; Paragraph 33) caused by predictions (Plurality of features or Parameters; See Fig 3a) of the upstream model (First model) and based on the one or more system-level errors (values) caused by predictions of the (Second Model) downstream model.
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:
Determining the scope and contents of the prior art.
Ascertaining the differences between the prior art and the claims at issue.
Resolving the level of ordinary skill in the pertinent art.
Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 2, 10, 17, 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Burton in view of Godfrey et al. (US 20200279192 A1, hereinafter referred to as Godfrey).
Regarding Claim 2, Burton teaches a method for determining confidence (Metric) for a system having two or more cascaded models (Two or more models) according to claim 1 (and thus the rejection of claim 1 is incorporated).
Burton teaches further comprising: evaluating confidence (Metric) of the system based on the generated confidence interval (Averaged model); Burton does not teach and adjusting the system to enhance precision of the system.
Godfrey teaches and adjusting the system to enhance precision of the system (See Fig.
6; “and retrain the first machine learning model when the accuracies of the classifications determined by the second machine learning model do not conform with the probability distribution of the corresponding assessment values determined by the first machine learning model.”, “deploy the retrained first machine learning model to the client electronic device.”; The new deployed retrained model enhances the precision of the system by being more accurate at predicating errors.).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Godfrey to evaluate confidence of the system based on the generated confidence interval and adjusting the system using the confidence intervals to enhance precision to increase the accuracy and precision of predicating errors in a machine learning system or method.
Regarding Claim 10, Burton teaches a computer system for determining confidence (metric) for a system having two or more cascaded models (Two or more models) according to claim 9 (and thus the rejection of claim 9 is incorporated).
Burton teaches further comprising: evaluating confidence (Metric) of the system based on the generated confidence interval (Averaged model); Burton does not teach and adjusting the system to enhance precision of the system.
Godfrey teaches and adjusting the system to enhance precision of the system (See Fig.
6; “and retrain the first machine learning model when the accuracies of the classifications determined by the second machine learning model do not conform with the probability distribution of the corresponding assessment values determined by the first machine learning
model.”, “deploy the retrained first machine learning model to the client electronic device.”; The new deployed retrained model enhances the precision of the system by being more accurate at predicting errors.).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Godfrey to evaluate confidence of the system based on the generated confidence interval and adjusting the system using the confidence intervals to enhance precision to increase the accuracy and precision of predicating errors in a machine learning system or method.
Regarding claim 17, Burton teaches a method for determining confidence (Metric) for a system having two or more cascaded models, the method comprising: generating a confidence interval (Averaged Model) that indicates a confidence ( Metric) for the system based on one or more system-level errors (Values) caused by predictions of an upstream model (first model) of the two or more cascaded models and based on the one or more system-level errors (Values) caused by predictions of a downstream model (second model) of the two or more cascaded models without using end to end system level data (Burton teaches not using end to end system data as the inputted data is not the outputted confidence interval nor the metric of probability of disease);
Burton does not teach evaluating and adjusting the system to enhance precision of the system.
Godfrey teaches evaluating and adjusting the system to enhance precision of the system (See Fig. 6; “and retrain the first machine learning model when the accuracies of the classifications determined by the second machine learning model do not conform with the
probability distribution of the corresponding assessment values determined by the first machine learning model.”, “deploy the retrained first machine learning model to the client electronic device.”).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Godfrey to evaluate confidence of the system based on the generated confidence interval without using end-to-end data and adjusting the system using the confidence intervals to enhance precision to increase the accuracy and precision of predicating errors in a machine learning system or method.
Regarding claim 19, most of the limitations of this claim have been noted in the rejection of claim 17 (and thus the rejection of claim 17 are incorporated).
Burton teaches wherein generating the confidence interval (Averaged Model) further comprises: estimating one or more empirical quantiles (“estimate one or more values associated with a presence of an expected disease state or condition”; See paragraph 45) at a predefined probability level (The predefined probability level in Burtons cases is the probability level of disease.).
Regarding claim 20, most of the limitations of this claim have been noted in the rejections of claim 17 and claim 19 (and thus the rejection of claim 17 and claim 19 are incorporated.).
Burton teaches wherein generating the confidence interval (Averaged Model) further comprises: determining empirical error distribution (outlier detection analysis; See paragraph 45; Outlier detection analysis would detect error and error rate in a given dataset.) for the first validation data set (First Biophysical signal Data Set) and the second validation data set (Second biophysical signal Data set).
Claims 3 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Burton in view of Haynes (US 20210046954 A1).
Regarding Claim 3, Burton teaches a method for determining confidence (Metric) for a system having two or more cascaded models (Two or more models) according to claim 1 (and thus the rejection of claim 1 is incorporated).
Regarding claim 3, Burton does not teach wherein the upstream model (First Model) comprises an object detection model and wherein the downstream model (Second Model) comprises a classification model.
Haynes teaches wherein the upstream model (Detection Component; See paragraph 86) comprises an object detection model (Object Detection models; See paragraph 86) and wherein the downstream model (classification component; See paragraph 86) comprises a classification model (Classification Models; See paragraph 86).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Haynes by using one model to detect and another to classify data for determining confidence to increase the accuracy and precision of predicating errors in a machine learning system.
Regarding Claim 11, Burton teaches a system for determining confidence (metric) for a system having two or more cascaded models (Two or more models) according to claim 9 (and thus the rejection of claim 9 is incorporated).
The method of claim 11, Burton does not teach wherein the upstream model (First Model) comprises an object detection model and wherein the downstream model (Second Model) comprises a classification model.
Haynes teaches wherein the upstream model (Detection Component; See paragraph 86) comprises an object detection model (Object Detection models; See paragraph 86) and wherein the downstream model (classification component; See paragraph 86) comprises a classification model (Classification Models; See paragraph 86).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Haynes by using one model to detect and another to classify data for determining confidence to increase the accuracy and precision of predicating errors in a machine learning system.
Claims 7-8, 15-16 are rejected under 35 U.S.C. 103 as being unpatentable over Burton in view of Hoogerwerf et al. (US 20210046954 A1, hereinafter referred to as Hoogerwerf).
Regarding claim 7, most of the limitations of this claim have been noted in the rejections of claim 1, claim 4, and claim 5 (and thus the rejections of claim 1, claim 4, and claim 5 are incorporated).
Burton does not teach wherein determining empirical error distribution further comprises: determining cluster-level error distribution.
Hoogerwerf teaches wherein determining empirical error distribution further comprises: determining cluster-level error distribution (See paragraph 13; “For example, grouping instances from a test dataset into feature clusters based on correlation measures between features and identified output errors of the machine learning system, the model evaluation system can provide tools and functionality to enable an individual to identify groupings of instances based on corresponding features”).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Hoogerwerf to use Cluster-level Error distribution, clustering similar errors, to increase the accuracy and precision of predicating errors in a machine learning system or method.
Regarding claim 8 most of the limitations of this claim have been noted in the rejection of claim 1 (and thus the rejections of claim 1 are incorporated)
Hoogerwerf teaches wherein generating the confidence interval (Features; paragraph 23; “This may include confidence scores”) is further configured to: generate the confidence interval (Features; paragraph 23; “This may include confidence scores”) using a split conformal prediction technique (Hoogerwerf splits the data into training data (Training Dataset; See paragraph 21) and test data (Test dataset; See paragraph 21), then clusters similar data based on predictions and similarities between the data so it can (See paragraph 23; “Using this data, systems described herein can describe errors with respect to system evidence rathe r than just content of an input.”)).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Hoogerwerf to use a split conformal prediction technique to increase the accuracy and precision of predicating errors in a machine learning system or method.
Regarding claim 15, most of the limitations of this claim have been noted in the rejections of claim 9, claim 12, and claim 13 (and thus the rejections of claim 9, claim 12, and claim 13 are incorporated).
Burton does not teach wherein determining empirical error distribution further comprises: determining cluster-level error distribution.
Hoogerwerf teaches wherein determining empirical error distribution further comprises: determining cluster-level error distribution (See paragraph 13; “For example, grouping instances from a test dataset into feature clusters based on correlation measures between features and identified output errors of the machine learning system, the model evaluation system can provide tools and functionality to enable an individual to identify groupings of instances based on corresponding features”).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Hoogerwerf to use Cluster-level Error distribution, clustering similar errors, to increase the accuracy and precision of predicating errors in a machine learning system or method.
Regarding claim 16 most of the limitations of this claim have been noted in the rejection of claim 9 (and thus the rejections of claim 9 are incorporated),
Hoogerwerf teaches wherein the machine learning system configured to generate the confidence interval (Features; paragraph 23; “This may include confidence scores”) is further configured to: generate the confidence interval (Features) using a split conformal prediction technique (Hoogerwerf splits the data into training data (Training Dataset; See paragraph 21) and test data (Test dataset; See paragraph 21), then clusters similar data based on predictions and similarities between the data so it can (See paragraph 23; “Using this data, systems described herein can describe errors with respect to system evidence rather than just content of an input.”)).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Hoogerwerf to using a split conformal prediction technique to increase the accuracy and precision of predicating errors in a machine learning system or method.
Claims 6 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Burton in view of Godfrey and in Further view of Hoogerwerf.
Regarding Claim 6 most of the limitations of this claim have been noted in the rejection of claim 1, claim 4, and claim 5 (and thus the rejections of claim 1, claim 4, and claim 5 are incorporated).
Burton does not teach grouping a plurality of data points in the first validation data set (First Biophysical Signal Data Set) and the second validation data set (Second Biophysical Signal Data Set) into two or more clusters based on similarity of intermediate features (Plurality of features or parameters), wherein the intermediate features comprise intermediate output from the upstream model (First model) and input to the downstream model (Second model).
Godfrey teaches a plurality of data points in the first validation data set (First assessment values; See Fig. 6) and the second validation data set (Second assessment values; See Fig 6.), wherein the intermediate features (classifications; See Fig. 6) comprise intermediate output (assessment values determined by the first machine learning model See Fig. 6) from the upstream model (First machine learning model; See Fig. 6) and input to the downstream model (second machine learning model ;See Fig.6 ; “apply the assessment values determined by the first machine learning model to a second machine learning model.”).
The combination of Burton and Godfrey teaches wherein the plurality of data points in the first and second data sets wherein the intermediate features comprise an intermediate output that is inputted into the downstream model. The combination of Burton and Godfrey does not teach grouping a plurality of data points in the first validation data set and the second validation data set into two or more clusters based on similarity of intermediate features.
Hoogerwerf discloses grouping a plurality of data points (“for example, grouping instances from a test dataset into feature clusters based on correlation measures between features and identified output errors of the machine learning system, the model evaluation system can provide tools and functionality to enable an individual to identify groupings of instances based on corresponding features”) in the first validation data set (Training Data) and the second validation data set (Test data) into two or more clusters (clusters) based on similarity of intermediate features (Features; See paragraph 13; “groupings of instances based on corresponding features”).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Godfrey and Hoogerwerf to group a plurality of data points in the first validation data set and the second validation data set into two or more clusters based on similarity of intermediate features, wherein the intermediate features comprise intermediate output from the upstream model and input to the downstream model to enhance the precision of the system.
Regarding Claim 14 most of the limitations of this claim have been noted in the rejection of claim 13, claim 12, and claim 9 (and thus the rejections of claim 13, claim 12, and claim 9 are incorporated).
Burton does not teach grouping a plurality of data points in the first validation data set (First Biophysical Signal Data Set) and the second validation data set (Second Biophysical Signal Data Set) into two or more clusters based on similarity of intermediate features (Plurality of features or parameters), wherein the intermediate features comprise intermediate output from the upstream model (First model) and input to the downstream model (Second model).
Godfrey teaches a plurality of data points in the first validation data set (First assessment values; See Fig. 6) and the second validation data set (Second assessment values; See Fig 6.), wherein the intermediate features (classifications; See Fig. 6) comprise intermediate output (assessment values determined by the first machine learning model See Fig. 6) from the upstream model (First machine learning model; See Fig. 6) and input to the downstream model (second machine learning model; See Fig.6; “apply the assessment values determined by the first machine learning model to a second machine learning model.”).
The combination of Burton and Godfrey teaches wherein the plurality of data points in the first and second data sets wherein the intermediate features comprise an intermediate output that is inputted into the downstream model. The combination of Burton and Godfrey does not teach grouping a plurality of data points in the first validation data set and the second validation data set into two or more clusters based on similarity of intermediate features.
Hoogerwerf discloses grouping a plurality of data points (“for example, grouping instances from a test dataset into feature clusters based on correlation measures between features and identified output errors of the machine learning system, the model evaluation system can provide tools and functionality to enable an individual to identify groupings of instances based on corresponding features”) in the first validation data set (Training Data) and the second validation data set (Test data) into two or more clusters (clusters) based on similarity of intermediate features (Features; See paragraph 13; “groupings of instances based on corresponding features”).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton to incorporate the teachings of Godfrey and Hoogerwerf to group a plurality of data points in the first validation data set and the second validation data set into two or more clusters based on similarity of intermediate features, wherein the intermediate features comprise intermediate output from the upstream model and input to the downstream model to enhance the precision of the system.
Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over Burton in view of Godfrey and in Further view of Haynes.
Regarding Claim 18, Burton and Godfrey teach a method for determining confidence (Metric) for a system having two or more cascaded models (Two or more models) according to claim 17 (and thus the rejection of claim 17 is incorporated).
The method of claim 18, Burton does not teach wherein the upstream model (First Model) comprises an object detection model and wherein the downstream model (Second Model) comprises a classification model.
Haynes teaches wherein the upstream model (Detection Component; See paragraph 86) comprises an object detection model (Object Detection models; See paragraph 86) and wherein the downstream model (classification component; See paragraph 86) comprises a classification model (Classification Models; See paragraph 86).
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention to have modified Burton and Godfrey to incorporate the teachings of Haynes to increase the accuracy and precision of predicating errors in a machine learning system by using one model to detect and another to classify data for determining confidence.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRETT DAVID KOLB JR whose telephone number is (571)270-0751. The examiner can normally be reached Monday-Friday.
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, Cesar Paula can be reached at (571) 272-4128. 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.
/BRETT DAVID KOLB JR/Examiner, Art Unit 2145
/CHAU T NGUYEN/Primary Examiner, Art Unit 2145
Prosecution Insights