US 12,585,999 B2 · Method and system for calibrating machine learning models in fully homomorphic encryption applications

Abstract

The present disclosure provides a technique to evaluate encrypted Machine Learning (ML) models. Conventional methods are unable to provide a holistic approach to evaluate encrypted ML models. Initially, the system receives an encrypted ML model. The ML model can be an unencrypted ML model trained with encrypted data or an encrypted ML model trained with encrypted data, or an encrypted ML model trained with unencrypted data. Further, a plurality of evaluation functions pertaining to the ML model to be calibrated are identified using a pattern matching technique. Further, an approximated function is generated for each of the plurality of evaluation functions using a corresponding approximation technique. After generating a plurality of approximated functions, an Expected Calibration Error (ECE) value is computed based on the plurality of approximated functions. Finally, the ML model is calibrated based on the computed ECE value. The ML model is perfectly calibrated if the computed ECE value is equal to zero.

Claims20

Dependent claims

A processor implemented method, the method comprising:

receiving, by one or more hardware processors, a Machine Learning (ML) model to be calibrated, wherein the ML model to be calibrated belongs to one of a) unencrypted ML model trained with encrypted data b) encrypted ML model trained with encrypted data and c) encrypted ML model trained with unencrypted data;
identifying, by the one or more hardware processors, a plurality of evaluation functions pertaining to the ML model, to be calibrated, using a pattern matching technique;
generating, by the one or more hardware processors, an approximated function for each of the plurality of evaluation functions using a corresponding approximation technique;
computing, by the one or more hardware processors, an Expected Calibration Error (ECE) value for the ML model based on a generated plurality of approximated functions using an approximated Broyden-Fletcher-Goldfarb-Shanno (BFGS) technique;
and
calibrating, by the one or more hardware processors, the ML model based on the ECE value.

3 more elements

A system comprising:

at least one memory storing programmed instructions;
one or more Input/Output (I/O) interfaces;
and one or more hardware processors operatively coupled to the at least one memory, wherein the one or more hardware processors are configured by the programmed instructions to:
receive a Machine Learning (ML) model to be calibrated, wherein the ML model to be calibrated belongs to one of a) unencrypted ML model trained with encrypted data b) encrypted ML model trained with encrypted data and c) encrypted ML model trained with unencrypted data;
identify a plurality of evaluation functions pertaining to the ML model, to be calibrated, using a pattern matching technique;
generate an approximated function for each of the plurality of evaluation functions using a corresponding approximation technique;
compute an Expected Calibration Error (ECE) value for the ML model based on a generated plurality of approximated functions using an approximated Broyden-Fletcher-Goldfarb-Shanno (BFGS) technique;
and
calibrate the ML model based on the ECE value.

6 more elements

Dependencies are inferred from the claim text.

Method and system for calibrating machine learning models in fully homomorphic encryption applications

Life of this patent

Method and system for calibrating machine learning models in fully homomorphic encryption applications

Life of this patent

Abstract

Claims20