Methods and systems for secure data analysis and machine learning

Abstract

Methods and systems for secure machine learning are disclosed. The methods include, by a processor: receiving a labeled dataset for use in training the machine learning model, and transmitting a first cluster of training data selected from the labeled dataset to a training device. The first cluster includes less than a threshold amount of data that is determined to prevent the training device from deriving information about the labeled dataset from the first cluster. The methods further include receiving a trained machine learning model from the training device, evaluating the trained machine learning to determine whether the trained machine learning model satisfies an evaluation criterion, and encrypting the trained machine learning model if the trained machine learning model satisfies the evaluation criterion. The encrypted machine learning model can then be deployed to make predictions using live data.