Machine Learning Predictors for Min-Entropy Estimation

Abstract

This study investigates the application of machine learning predictors for the estimation of min-entropy in random number generators (RNGs), a key component in cryptographic applications where accurate entropy assessment is essential for cybersecurity. Our research indicates that these predictors, and indeed any predictor that leverages sequence correlations, primarily estimate average min-entropy, a metric not extensively studied in this context. We explore the relationship between average min-entropy and the traditional min-entropy, focusing on their dependence on the number of target bits being predicted. Using data from generalized binary autoregressive models, a subset of Markov processes, we demonstrate that machine learning models (including a hybrid of convolutional and recurrent long short-term memory layers and the transformer-based GPT-2 model) outperform traditional NIST SP 800-90B predictors in certain scenarios. Our findings underscore the importance of considering the number of target bits in min-entropy assessment for RNGs and highlight the potential of machine learning approaches in enhancing entropy estimation techniques for improved cryptographic security.