Solution for machine learning system

1. A non-transitory computer-readable medium on which is stored program that, when executed by a computer, performs a method for estimating an uncertainty of a prediction generated by a machine learning system, the method comprising:

receiving first data;

training a first machine learning model component of a machine learning system with the received first data, the first machine learning model component is trained to generate a prediction;

generating an uncertainty estimate of the prediction; and

training a second machine learning model component of the machine learning system with second data, the second machine learning model component is trained to generate a calibrated uncertainty estimate of the prediction based on the prediction, the uncertainty estimate of the prediction, and an output of at least one anomaly detector.

2. The computer-readable medium of claim 1 , wherein the uncertainty estimate of the prediction is generated by one of the first machine learning model component, the second machine learning model component, or an external machine learning model component.

3. The computer-readable medium of claim 1 , wherein the anomaly detector is trained with the received second data for detecting deviation in the operational data.

4. The computer-readable medium of claim 1 , wherein the first machine learning model component is one of a denoising neural network, a generative adversarial network, a variational autoencoder, a ladder network, or a recurrent neural network.

5. The computer-readable medium of claim 1 , wherein the second machine learning model component is one of a denoising neural network, a generative adversarial network, a variational autoencoder, a ladder network, or a recurrent neural network.

6. The computer-readable medium of claim 1 , wherein the second data is out-of-distribution data.

7. The computer-readable medium of claim 6 , wherein the out-of-distribution data is generated by corrupting the first machine learning model component parameters and generating the out-of-distribution data by evaluating the corrupted first machine learning model component.

8. A system for estimating an uncertainty of a prediction generated by a machine learning system, the system is arranged to:

receive first data,

train a first machine learning model component of a machine learning system with the received first data, the first machine learning model component is trained to generate a prediction,

generate an uncertainty estimate of the prediction,

train a second machine learning model component of the machine learning system with second data, the second machine learning model component is trained to generate a calibrated uncertainty estimate of the prediction based on the prediction, the uncertainty estimate of the prediction, and an output of at least one anomaly detector.

9. The system of claim 8 , wherein the system is arranged to generate the uncertainty estimate of the prediction by one of the first machine learning model component, the second machine learning model component, or an external machine learning model component.

10. The system of claim 8 , wherein the system is arranged to train the anomaly detector with the received second data for detecting deviation in the operational data.

11. The system of claim 8 , wherein the first machine learning model component is one of a denoising neural network, a generative adversarial network, a variational autoencoder, a ladder network, or a recurrent neural network.

12. The system of claim 8 , wherein the second machine learning model component is one of a denoising neural network, a generative adversarial network, a variational autoencoder, a ladder network, or a recurrent neural network.

13. The system of claim 8 , wherein the second data is out-of-distribution data.

14. The system of claim 13 , wherein the out-of-distribution data is generated by corrupting the first machine learning model component parameters and generating the out-of-distribution data by evaluating the corrupted first machine learning model component.

15. A method for estimating an uncertainty of a prediction generated by a machine learning system, the method comprising:

receiving first data;

training a first machine learning model component of a machine learning system with the received first data, the first machine learning model component is trained to generate a prediction;

generating an uncertainty estimate of the prediction; and

training a second machine learning model component of the machine learning system with second data, the second machine learning model component is trained to generate a calibrated uncertainty estimate of the prediction based on the prediction, the uncertainty estimate of the prediction, and an output of at least one anomaly detector.

16. The method of claim 15 , wherein the uncertainty estimate of the prediction is generated by one of the first machine learning model component, the second machine learning model component, or an external machine learning model component.

17. The method of claim 15 , wherein the anomaly detector is trained with the received second data for detecting deviation in the operational data.

18. The method of claim 15 , wherein the first machine learning model component is one of a denoising neural network, a generative adversarial network, a variational autoencoder, a ladder network, or a recurrent neural network.

19. The method of claim 15 , wherein the second machine learning model component is one of a denoising neural network, a generative adversarial network, a variational autoencoder, a ladder network, or a recurrent neural network.

20. The method of claim 15 , wherein the second data is out-of-distribution data.

21. The method of claim 20 , wherein the out-of-distribution data is generated by corrupting the first machine learning model component parameters and generating the out-of-distribution data by evaluating the corrupted first machine learning model component.