INFORMATION TRANSMISSION METHOD AND APPARATUS

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 § 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. Claim(s) 1-11 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Vasseur et al., (US 2021/0158106 A1, hereinafter Vasseur). Regarding claims 1 and 9, taking claim 9 as exemplary: Vasseur shows: “An apparatus chip, comprising a processor and a memory, wherein the memory is configured to store a computer program, and the processor is configured to invoke and run the computer program stored in the memory,” (Paragraph [0036]: “The memory 240 comprises a plurality of storage locations that are addressable by the processor(s) 220 and the network interfaces 210 for storing software programs and data structures associated with the embodiments described herein. The processor 220 may comprise necessary elements or logic adapted to execute the software programs and manipulate the data structures 245. An operating system 242 (e.g., the Internetworking Operating System, or IOS®, of Cisco Systems, Inc., another operating system, etc.), portions of which are typically resident in memory 240 and executed by the processor(s), functionally organizes the node by, inter alia, invoking network operations in support of software processors and/or services executing on the device. These software processors and/or services may comprise a machine learning operations (ML Ops) process 248, as described herein, any of which may alternatively be located within individual network interfaces.” And in paragraph [0126]: “The foregoing description has been directed to specific embodiments. It will be apparent, however, that other variations and modifications may be made to the described embodiments, with the attainment of some or all of their advantages. For instance, it is expressly contemplated that the components and/or elements described herein can be implemented as software being stored on a tangible (non-transitory) computer-readable medium (e.g., disks/CDs/RAM/EEPROM/etc.) having program instructions executing on a computer, hardware, firmware, or a combination thereof. Accordingly, this description is to be taken only by way of example and not to otherwise limit the scope of the embodiments herein. Therefore, it is the object of the appended claims to cover all such variations and modifications as come within the true spirit and scope of the embodiments herein.”) “to perform: sending a first type of feedback information to a second network device,” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” And in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.”) “wherein the first type comprises one or more of the following: a model evaluation indicator, operation indication information, and a network performance indicator, wherein the model evaluation indicator indicates performance of a model, the operation indication information indicates an operation to be performed based on an analytics result of the model, the network performance indicator indicates a network indicator obtained based on the analytics result of the model, and the first type is used to request the feedback information;” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” And in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.”) “receiving the feedback information from the second network device; and determining the performance of the model based on the feedback information.” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” And in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.”) Regarding claims 2 and 10, taking claim 10 as exemplary: Vasseur shows the method and system of claims 1 and 9 as claimed and specified above. And Vasseur shows “wherein the processor is further configured to perform: receiving, an actual value of the model or the network performance indicator from a third network device; and wherein the determining, the performance of the model based on the feedback information comprises: determining, the performance of the model based on the actual value of the model or the network performance indicator and the feedback information.” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” And in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.”) Regarding claims 3 and 11, taking claim 11 as exemplary: Vasseur shows the method and system of claims 1 and 9 as claimed and specified above. And Vasseur shows “wherein the processor is further configured to perform: receiving, first indication information from the second network device, wherein the first indication information indicates a second type of the feedback information, wherein the second type of the feedback information is supported by the second network device; and determining, the first type based on the first indication information.” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” And in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.” And in paragraph [0116]: “feedback engine 416 may send instructions 420 to dynamically increase the data quality/fidelity of the telemetry data assessed by the machine learning model of wireless network assurance service 402. As would be appreciated, some machine learning algorithms are inherently tolerant to missing data. Accordingly, feedback engine 416 may also take into account the tolerance(s) of the algorithm to the data fidelity issues (e.g., missing data, spurious data, etc.), before altering the generation of the telemetry data for input to the model. To do so, engine 416 may first determine whether the data quality metrics from telemetry fidelity inspector 410 meet the maximum tolerance of the algorithm or overall system. If so, then feedback engine 416 may take no specific action. However, if the data fidelity level is below the specified tolerance of the machine learning algorithm, or when no tolerance is specified, feedback engine…”) Regarding claim 4: Vasseur shows: “An information transmission method comprising: receiving, by a second network device, a first type of feedback information from a first network device, wherein the first type comprises one or more of the following: a model evaluation indicator, operation indication information, and a network performance indicator, wherein the model evaluation indicator indicates performance of a model, the operation indication information indicates an operation to be performed based on an analytics result of the model, the network performance indicator indicates a network indicator obtained based on the analytics result of the model, and the first type is used to request the feedback information;” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” In paragraph [0116]: “feedback engine 416 may send instructions 420 to dynamically increase the data quality/fidelity of the telemetry data assessed by the machine learning model of wireless network assurance service 402. As would be appreciated, some machine learning algorithms are inherently tolerant to missing data. Accordingly, feedback engine 416 may also take into account the tolerance(s) of the algorithm to the data fidelity issues (e.g., missing data, spurious data, etc.), before altering the generation of the telemetry data for input to the model. To do so, engine 416 may first determine whether the data quality metrics from telemetry fidelity inspector 410 meet the maximum tolerance of the algorithm or overall system. If so, then feedback engine 416 may take no specific action. However, if the data fidelity level is below the specified tolerance of the machine learning algorithm, or when no tolerance is specified, feedback engine 416 may perform”, in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.” And in paragraph [0120]: “At step 1115, as detailed above, the service may detect unacceptable performance of the machine learning model. For example, assume that the model comprises an anomaly detector configured to detect anomalous behaviors of the monitored network. In such a case, unacceptable performance of the model may correspond to the anomaly detector detecting more than a threshold percentage of anomalies (e.g., within a certain amount of time) Likewise, unacceptable performance of the model may be measured in terms of precision, recall, false positives, false negatives, combinations thereof, or the like.”) “and sending, by the second network device, the feedback information to the first network device based on the first type of the feedback information.” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” And in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.”) Regarding claim 5: Vasseur shows the method of claim 4 as claimed and specified above. And Vasseur shows “wherein the sending, by the second network device, the feedback information to the first network device based on the first type of the feedback information comprises: determining, by the second network device, the feedback information based on the first type of the feedback information; and sending, by the second network device, the feedback information to the first network device.” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” In paragraph [0116]: “feedback engine 416 may send instructions 420 to dynamically increase the data quality/fidelity of the telemetry data assessed by the machine learning model of wireless network assurance service 402. As would be appreciated, some machine learning algorithms are inherently tolerant to missing data. Accordingly, feedback engine 416 may also take into account the tolerance(s) of the algorithm to the data fidelity issues (e.g., missing data, spurious data, etc.), before altering the generation of the telemetry data for input to the model. To do so, engine 416 may first determine whether the data quality metrics from telemetry fidelity inspector 410 meet the maximum tolerance of the algorithm or overall system. If so, then feedback engine 416 may take no specific action. However, if the data fidelity level is below the specified tolerance of the machine learning algorithm, or when no tolerance is specified, feedback engine 416 may perform”, in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.” And in paragraph [0120]: “At step 1115, as detailed above, the service may detect unacceptable performance of the machine learning model. For example, assume that the model comprises an anomaly detector configured to detect anomalous behaviors of the monitored network. In such a case, unacceptable performance of the model may correspond to the anomaly detector detecting more than a threshold percentage of anomalies (e.g., within a certain amount of time) Likewise, unacceptable performance of the model may be measured in terms of precision, recall, false positives, false negatives, combinations thereof, or the like.”) Regarding claim 6: Vasseur shows the method of claim 4 as claimed and specified above. And Vasseur shows “wherein the sending, by the second network device, the feedback information to the first network device based on the first type of the feedback information comprises: receiving, by the second network device, the feedback information from a fourth network device; and sending, by the second network device, the feedback information to the first network device.” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” In paragraph [0116]: “feedback engine 416 may send instructions 420 to dynamically increase the data quality/fidelity of the telemetry data assessed by the machine learning model of wireless network assurance service 402. As would be appreciated, some machine learning algorithms are inherently tolerant to missing data. Accordingly, feedback engine 416 may also take into account the tolerance(s) of the algorithm to the data fidelity issues (e.g., missing data, spurious data, etc.), before altering the generation of the telemetry data for input to the model. To do so, engine 416 may first determine whether the data quality metrics from telemetry fidelity inspector 410 meet the maximum tolerance of the algorithm or overall system. If so, then feedback engine 416 may take no specific action. However, if the data fidelity level is below the specified tolerance of the machine learning algorithm, or when no tolerance is specified, feedback engine 416 may perform”, in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.” And in paragraph [0120]: “At step 1115, as detailed above, the service may detect unacceptable performance of the machine learning model. For example, assume that the model comprises an anomaly detector configured to detect anomalous behaviors of the monitored network. In such a case, unacceptable performance of the model may correspond to the anomaly detector detecting more than a threshold percentage of anomalies (e.g., within a certain amount of time) Likewise, unacceptable performance of the model may be measured in terms of precision, recall, false positives, false negatives, combinations thereof, or the like.”) Regarding claim 7: Vasseur shows the method of claim 4 as claimed and specified above. And Vasseur shows “wherein the determining, by the second network device, the feedback information based on the first type of the feedback information comprises: receiving, by the second network device, an actual value of the model from a fifth network device; determining, by the second network device, the feedback information based on the first type and the actual value of the model; and sending, by the second network device, the feedback information to the first network device.” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” In paragraph [0116]: “feedback engine 416 may send instructions 420 to dynamically increase the data quality/fidelity of the telemetry data assessed by the machine learning model of wireless network assurance service 402. As would be appreciated, some machine learning algorithms are inherently tolerant to missing data. Accordingly, feedback engine 416 may also take into account the tolerance(s) of the algorithm to the data fidelity issues (e.g., missing data, spurious data, etc.), before altering the generation of the telemetry data for input to the model. To do so, engine 416 may first determine whether the data quality metrics from telemetry fidelity inspector 410 meet the maximum tolerance of the algorithm or overall system. If so, then feedback engine 416 may take no specific action. However, if the data fidelity level is below the specified tolerance of the machine learning algorithm, or when no tolerance is specified, feedback engine 416 may perform”, in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.” And in paragraph [0120]: “At step 1115, as detailed above, the service may detect unacceptable performance of the machine learning model. For example, assume that the model comprises an anomaly detector configured to detect anomalous behaviors of the monitored network. In such a case, unacceptable performance of the model may correspond to the anomaly detector detecting more than a threshold percentage of anomalies (e.g., within a certain amount of time) Likewise, unacceptable performance of the model may be measured in terms of precision, recall, false positives, false negatives, combinations thereof, or the like.”) Regarding claim 8: Vasseur shows the method of claim 4 as claimed and specified above. And Vasseur shows “further comprising: sending, by the second network device, first indication information to the first network device, wherein the first indication information indicates a second type of the feedback information, and the second type of the feedback information is supported by the second network device.” (Paragraph [0115]: “Referring yet again to FIG. 4, a final component of architecture 400 may be feedback engine 416 that is configured to implement a feedback control loop with the network monitoring service (e.g., any of services 402-406), so as to address any model performance issues attributable to a lack of data fidelity of its input telemetry data. In general, this may entail curating telemetry data on the fly, reconstructing spurious data, or even dynamically requesting additional telemetry data on the fly from the telemetry collection mechanism in the monitored network. For example, if ML Ops service 408 determines that the performance of the model of wireless network assurance service 402 is unacceptable due to a data fidelity issue, feedback engine 416 may send one or more instructions 420 back to service 402 to alter how service 402 collects or analyzes telemetry data from the monitored network.” In paragraph [0116]: “feedback engine 416 may send instructions 420 to dynamically increase the data quality/fidelity of the telemetry data assessed by the machine learning model of wireless network assurance service 402. As would be appreciated, some machine learning algorithms are inherently tolerant to missing data. Accordingly, feedback engine 416 may also take into account the tolerance(s) of the algorithm to the data fidelity issues (e.g., missing data, spurious data, etc.), before altering the generation of the telemetry data for input to the model. To do so, engine 416 may first determine whether the data quality metrics from telemetry fidelity inspector 410 meet the maximum tolerance of the algorithm or overall system. If so, then feedback engine 416 may take no specific action. However, if the data fidelity level is below the specified tolerance of the machine learning algorithm, or when no tolerance is specified, feedback engine 416 may perform”, in paragraph [0119]: “FIG. 11 illustrates an example simplified procedure for assessing data quality in a machine learning-based network assurance system or other system that monitors a network, in accordance with one or more embodiments described herein. For example, a non-generic, specifically configured device may perform procedure 1100 by executing stored instructions, to provide a service to one or more networks. The procedure 1100 may start at step 1105, and continues to step 1110, where, as described in greater detail above, the service may compute a data fidelity metric for network telemetry data used by a machine learning model to monitor a computer network. For example, such a model may be executed by a network assurance service/system, a system/service that uses the model to classify devices in the network by device type, or the like. In various embodiments, the computed data fidelity metric may be based in part on a sampling frequency at which the network telemetry data is collected in the network, based in part on missing values in the telemetry data, or based in part on spurious data in the telemetry data.” And in paragraph [0120]: “At step 1115, as detailed above, the service may detect unacceptable performance of the machine learning model. For example, assume that the model comprises an anomaly detector configured to detect anomalous behaviors of the monitored network. In such a case, unacceptable performance of the model may correspond to the anomaly detector detecting more than a threshold percentage of anomalies (e.g., within a certain amount of time) Likewise, unacceptable performance of the model may be measured in terms of precision, recall, false positives, false negatives, combinations thereof, or the like.”) Conclusion The prior art made of record and not relied upon is considered pertinent to applicant's disclosure: Froehlich et al., (US 2020/0022006 A1), part of the prior art made of record, teaches the using a model to evaluate performance of a network of claims 1, 4, and 9 in paragraph [0038] through the use of a classification model with rules that monitors call dropping in a wireless network. Bejerano et al., (US 2017/0347279 A1), part of the prior art made of record, teaches the using a model to evaluate performance of a network of claims 1, 4, and 9 in paragraph [0006] through the use of a feedback information in a network to determine switching a modes for wireless devices. Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHANE D WOOLWINE whose telephone number is (571)272-4138. The examiner can normally be reached M-F 9:30-6:00 PM. 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, MIRANDA HUANG can be reached at (571) 270-7092. 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. SHANE D. WOOLWINE Primary Examiner Art Unit 2124 /SHANE D WOOLWINE/Primary Examiner, Art Unit 2124