Passive inferencing of signal following in multivariate anomaly detection
Systems, methods, and other embodiments associated with passive inferencing of signal following in multivariate anomaly detection are described. In one embodiment, a method for inferencing signal following in a machine learning (ML) model includes calculating an average standard deviation of measured values of time series signals in a set of time series signals; training the ML model to predict values of the signals; predicting values of each of the signals with the trained ML model; generating a time series set of residuals between the predicted values and the measured values; calculating an average standard deviation of the sets of residuals; determining that signal following is present in the trained ML model where a ratio of the average standard deviation of measured values to the average standard deviation of the sets of residuals exceeds a threshold; and presenting an alert indicating the presence of signal following in the trained ML model.
1 . A computer-implemented method for inferencing signal following in a machine learning model, the method comprising:
calculating an average standard deviation of measured values of time series signals in a set of more than one time series signal;
training the machine learning model to predict values of the time series signals, by:
feeding an observation interval of the time series signals that represents a normal operating state into the machine learning model for the training, and
configuring the machine learning model to generate for individual signals of the time series signals an output signal value that would be expected in the normal operating state based on input signal values and correlations among the time series signals other than the individual signals;
predicting values of each of the time series signals with the trained machine learning model;
generating a time series set of residuals between the predicted values and the measured values for each of the time series signals;
calculating an average standard deviation of the sets of residuals;
determining that signal following is present in the trained machine learning model where a ratio of the average standard deviation of measured values to the average standard deviation of the sets of residuals exceeds a threshold; and
presenting an alert indicating the presence of signal following in the trained machine learning model.
2 . The computer-implemented method of claim 1 , further comprising calculating the ratio by dividing the average standard deviation of measured values by the average standard deviation of the sets of residuals, wherein the threshold exceeded by the ratio is between 1 and 1.5.
3 . The computer-implemented method of claim 1 , further comprising retrieving a quantitative degree of following for the machine learning model based on a value of the ratio.
4 . The computer-implemented method of claim 1 , wherein the training of the machine learning model and predicting of values occurs only once when inferencing signal following in the machine learning model.
5 . The computer-implemented method of claim 1 , wherein the alert includes the ratio and a recommendation of a technique to mitigate the signal following in the machine learning model selected from increasing a number of training vectors in the machine learning model, filtering signal inputs to the machine learing model to reduce noise, and increasing a number of signals.
6 . The computer-implemented method of claim 1 , wherein the machine learning model is a multivariate state estimation technique model.
7 . The computer-implemented method of claim 1 , wherein the machine learning model is a non-linear non-parametric regression model.
8 . A non-transitory computer-readable medium that includes stored thereon computer-executable instructions that when executed by at least a processor of a computer cause the computer to:
calculate an average standard deviation of measured values of time series signals in a set of more than one time series signal;
train a machine learning model to predict values of the time series signals, by:
feeding an observation interval of the time series signals that represents a normal operating state into the machine learning model for the training, and
configuring the machine learning model to generate for individual signals of the time series signals an output signal value that would be expected in the normal operating state based on input signal values and correlations among the time series signals other than the individual signals;
predict values of each of the time series signals with the trained machine learning model;
generate a time series set of residuals between the predicted values and the measured values for each of the time series signals;
calculate an average standard deviation of the sets of residuals;
determine that signal following is present in the trained machine learning model where a ratio of the average standard deviation of measured values to the average standard deviation of the sets of residuals exceeds a threshold; and
present an alert indicating the presence of signal following in the trained machine learning model.
9 . The non-transitory computer-readable medium of claim 8 , wherein the instructions further cause the computer to calculate the ratio by dividing the average standard deviation of measured values by the average standard deviation of the sets of residuals, and wherein the threshold exceeded by the ratio is between 1 and 1.5.
10 . The non-transitory computer-readable medium of claim 8 , wherein the instructions further cause the computer to retrieve a quantitative degree of following for the machine learning model based on a value of the ratio.
11 . The non-transitory computer-readable medium of claim 8 , wherein the training of the machine learning model and predicting of values occurs only once when inferencing signal following in the machine learning model.
12 . The non-transitory computer-readable medium of claim 8 , wherein the alert includes the ratio and a recommendation of a technique to mitigate the signal following in the machine learning model selected from increasing a number of training vectors in the machine learning model, filtering signal inputs to the machine learning model to reduce noise, and increasing a number of signals.
13 . The non-transitory computer-readable medium of claim 8 , wherein the machine learning model is a multivariate state estimation technique model.
14 . The non-transitory computer-readable medium of claim 8 , wherein the machine learning model is a non-linear non-parametric regression model.
15 . A computing system comprising:
a processor;
a memory operably connected to the processor;
a non-transitory computer-readable medium operably connected to the processor and memory and storing computer-executable instructions that when executed by at least a processor of a computer cause the computing system to:
calculate an average standard deviation of measured values of time series signals in a set of more than one time series signal;
train a machine learning model to predict values of the time series signals, by:
feeding an observation interval of the time series signals that represents a normal operating state into the machine learning model for the training, and
configuring the machine learning model to generate for individual signals of the time series signals an output signal value that would be expected in the normal operating state based on input signal values and correlations among the time series signals other than the individual signals;
predict values of each of the time series signals with the trained machine learning model;
generate a time series set of residuals between the predicted values and the measured values for each of the time series signals;
calculate an average standard deviation of the sets of residuals;
determine that signal following is present in the trained machine learning model where a ratio of the average standard deviation of measured values to the average standard deviation of the sets of residuals exceeds a threshold; and
present an alert indicating the presence of signal following in the trained machine learning model.
16 . The computing system of claim 15 , wherein the instructions further cause the computing system to calculate the ratio by dividing the average standard deviation of measured values by the average standard deviation of the sets of residuals, and wherein the threshold exceeded by the ratio is between 1 and 1.5.
17 . The computing system of claim 15 , wherein the instructions further cause the computing system to retrieve a quantitative degree of following for the machine learning model based on a value of the ratio.
18 . The computing system of claim 15 , wherein the alert includes the ratio and a recommendation of a technique to mitigate the signal following in the machine learning model selected from increasing a number of training vectors in the machine learning model, filtering signal inputs to the machine learning model to reduce noise, and increasing a number of signals.
19 . The computing system of claim 15 , wherein the machine learning model is a multivariate state estimation technique model.
20 . The computing system of claim 15 , wherein the machine learning model is a non-linear non-parametric regression model.