Random numbers are incredibly important in a variety of fields, and the need for their validation remains important for safety. A Quantum Random Number Generator (QRNG) can theoretically generate truly random numbers, however their quality still needs to be thoroughly validated. Generally, the task of validating random numbers has been delegated to different statistical tests such as the tests from the NIST Statistical Test Suite (STS), which are often slow and only perform one test at a time. Our work presents a deep learning model utilizing the Transformer architecture that 1) performs multiple NIST STS tests at once, and 2) runs much faster. This model outputs multilabel classification results on passing these statistical tests. We performed a thorough hyper-parameter optimization to converge on the best possible model and as a result, achieved a high degree of accuracy with a Macro F1-score of above 0.96. We also compared this model to a conventional deep learning method (Long Short Term Memory Recurrent Neural Networks) to quantify randomness and showed our model achieved similar performances while being much more efficient and scalable. The high performance and efficiency of this Transformer-based deep learning model showed that it can be a viable replacement for the NIST STS for validating random numbers.
Random numbers are incredibly important in a variety of fields, and the need for their validation remains important for safety. A Quantum Random Number Generator (QRNG) can theoretically generate truly random numbers, however their quality still needs to be thoroughly validated. Generally, the task of validating random numbers has been delegated to different statistical tests such as the tests from the NIST Statistical Test Suite (STS), which are often slow and only perform one test at a time. Our work presents a deep learning model utilizing the Transformer architecture that 1) performs multiple NIST STS tests at once, and 2) runs much faster. This model outputs multilabel classification results on passing these statistical tests. We performed a thorough hyper-parameter optimization to converge on the best possible model and as a result, achieved a high degree of accuracy with a Macro F1-score of above 0.96. We also compared this model to a conventional deep learning method (Long Short Term Memory Recurrent Neural Networks) to quantify randomness and showed our model achieved similar performances while being much more efficient and scalable. The high performance and efficiency of this Transformer-based deep learning model showed that it can be a viable replacement for the NIST STS for validating random numbers.
The random number generator (RNG) plays an essential role in lots of applications such as cryptographic. Unlike classical random number generators, quantum random number generators (QRNG) can in principle generate true random numbers because it is based on the inherent uncertainty of the quantum property. However, the presence of the environmental noise or the imperfect characteristics of the device can compromise the integrity of QRNG. To judge the quality of QRNG, we propose to apply the transformer based machine learning (ML) model to quantify the randomness of quantum random numbers. Through proper training, our transformer based ML model is able to predict randomness of given quantum random number sequence very closely to the maximum guess probability of its theoretical limit. Moreover, we also do comparisons between our present transformer based model with other ML model, such as the long short term memory network (LSTM) based model. Simulation results show that our present model can not only present a higher prediction accuracy, but also cost much less time and exhibit excellent performance in work efficiency. Furthermore, we also compare our present transformer based machine learning model with the traditional randomness detection standard, NIST-test-suite, and show advantages of our method.
This paper explores a novel randomness evaluation method for data produced by quantum random number generators (QRNGs), leveraging quantum mechanics to achieve highly random data. We employ neural networks and machine learning techniques to analyze the operational principles of QRNGs, enabling the test suites to assess the generators based on their predictability scores. Our findings demonstrate that our model’s ability to predict outcomes surpasses that of comparable approaches. Additionally, we discuss the optimal timing for conducting these tests—specifically, analyzing the raw output from QRNGs before processing through an extractor yields the best performance.
Quantum Random Number Generators (QRNGs) exploit intrinsic quantum mechanical randomness and are promising candidates for higher-quality randomness than classical RNGs. However, QRNGs on Noisy Intermediate-Scale Quantum (NISQ) platforms are constrained by decoherence, gate and measurement errors, and limited circuit depth, and the resulting impact on statistical randomness is not yet systematically characterized. This work introduces parameterized quantum random number generator (PQRNG) architectures built from parameterized quantum circuits (PQCs) to improve tunability and expressive power under realistic noise. We study three architectures, PQC-H-CH, H-PQC-CH, and H-CH-PQC, and evaluate their randomness under Transpiler optimization and circuit-level error mitigation as preprocessing steps, using NIST SP 800-22 for statistical validation. Utilizing default NIST settings, PQC-H-CH attains the largest number of fully passing preprocessing configurations (126 combinations passing all 15 tests), compared with 110 for H-PQC-CH and 69 for H-CH-PQC, indicating strong robustness across Transpiler optimization and mitigation settings, and we find that preprocessing choices influence randomness quality more strongly than adjusting NIST test parameters. Overall, these results demonstrate that the proposed approach provides a solid foundation for developing more reliable and practical QRNGs using PQCs in NISQ devices.
Quantum random number generators (QRNGs) offer a fundamentally different approach to random number generation compared to classical methods, leveraging the inherent unpredictability of quantum mechanical processes. However, the widespread adoption of QRNGs is currently hindered by a lack of standardized verification procedures and effective multi-model fusion strategies. This paper addresses these challenges by proposing a comprehensive framework for the multi-model validation and fusion of QRNGs. The framework integrates multiple quantum physical and information-theoretic randomness sources, employing statistical modeling for rigorous security assessments. Specifically, we introduce a novel approach to quantifying the entropy and bias of QRNG outputs across various models. The core contribution lies in establishing a systematic methodology for evaluating QRNG security and presenting a viable strategy for combining the strengths of different QRNG technologies. Our analysis highlights the importance of diverse validation techniques and the potential for improved QRNG performance through model fusion. The proposed framework provides a foundation for more reliable and trustworthy QRNG implementations, crucial for applications demanding high-quality randomness, such as cryptography and simulation.
Quantum random number generators (QRNG) provide quality random numbers, which are essential for cryptography by utilizing the unpredictable nature of quantum mechanics.Advancements in quantum optics made multiple different architectures for these possible.As part of a project aiming to realize a QRNG service, we developed a system capable of providing real-time monitoring and long term data collection while still fulfilling regular processing duties for these devices.In most cases, hardware validation is done by simply running a battery of statistical tests on the final output.Our goal, however, was to create a system allowing more flexible use of these tests, realizing a tool that can also prove useful during the construction of our entropy source for detecting and correcting unique imperfections.We tested this flexibility and the system's ability to adequately perform the required tasks with simulated sources while further examining the usability of available verification tools within this new custom framework.
Random prime number generation is crucial for the implementation of several encryption and signature protocols in cryptographic applications. Recently, Quantum Random Number Generators (QRNGs) have emerged as a solution to obtain information-theoretically secure randomness, setting them apart from algorithmic generators. Nonetheless, their output is still prone to deterioration under a flawed implementation and thus requires continuous statistical validation. This process is uniquely challenging for prime number generators since they cannot be submitted to the traditional testing suites, which are primarily designed to test uniform pseudorandom sources. Here, we enable direct statistical testing of prime Random Number Generators (RNGs) by extending a validation framework based on an equiprobable binning of the prime distribution with a predictive machine learning model, which can independently learn correlations in the prime output. The validity of this framework is then verified through extensive cryptanalysis of a prime QRNG based on quadrature measurements of the vacuum state. The developed model successfully identifies flawed configurations of the prime QRNG for output lengths up to 128 bits. This assessment was additionally extended to a purely classical scheme derived from electronic noise measurements. Although no increase in the model’s predictive capacity is seen, our testing framework can nonetheless demonstrate that this classical source produces a conclusively biased prime distribution. In turn, the implemented QRNG remains resilient against this type of cryptanalysis.