Statistical Quality and Reproducibility of Pseudorandom Number Generators in Machine Learning technologies

• URL: http://arxiv.org/abs/2507.03007v1
• Date: Wed, 02 Jul 2025 09:38:00 GMT
• Title: Statistical Quality and Reproducibility of Pseudorandom Number Generators in Machine Learning technologies
• Authors: Benjamin A. Antunes,
• Abstract summary: We compare the statistical quality of PRNGs used in ML frameworks against their original C implementations.<n>Our findings challenge claims of statistical robustness, revealing that even generators labeled ''crush-resistant'' (e.g., PCG, Philox) may fail certain statistical tests.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Machine learning (ML) frameworks rely heavily on pseudorandom number generators (PRNGs) for tasks such as data shuffling, weight initialization, dropout, and optimization. Yet, the statistical quality and reproducibility of these generators-particularly when integrated into frameworks like PyTorch, TensorFlow, and NumPy-are underexplored. In this paper, we compare the statistical quality of PRNGs used in ML frameworks (Mersenne Twister, PCG, and Philox) against their original C implementations. Using the rigorous TestU01 BigCrush test suite, we evaluate 896 independent random streams for each generator. Our findings challenge claims of statistical robustness, revealing that even generators labeled ''crush-resistant'' (e.g., PCG, Philox) may fail certain statistical tests. Surprisingly, we can observe some differences in failure profiles between the native and framework-integrated versions of the same algorithm, highlighting some implementation differences that may exist.

Related papers

• Satori-SWE: Evolutionary Test-Time Scaling for Sample-Efficient Software Engineering [51.7496756448709]
  Language models (LMs) perform well on coding benchmarks but struggle with real-world software engineering tasks.<n>Existing approaches rely on supervised fine-tuning with high-quality data, which is expensive to curate at scale.<n>We propose Test-Time Scaling (EvoScale), a sample-efficient method that treats generation as an evolutionary process.
  arXiv  Detail & Related papers  (2025-05-29T16:15:36Z)
• Efficient Quality Estimation of True Random Bit-streams [5.441027708840589]
  This paper reports the implementation and characterization of an on-line procedure for the detection of anomalies in a true random bit stream. The experimental validation of the approach is performed upon the bit streams generated by a quantum, silicon-based entropy source.
  arXiv  Detail & Related papers  (2024-09-09T12:09:17Z)
• Transformer models as an efficient replacement for statistical test suites to evaluate the quality of random numbers [0.0]
  We present a deep learning model that performs multiple NIST STS tests at once and runs much faster. This model outputs multi-label classification results on passing these statistical tests. We also compared this model to a conventional deep learning method to quantify randomness and showed our model achieved similar performances.
  arXiv  Detail & Related papers  (2024-05-06T23:36:03Z)
• To what extent are multiple pendulum systems viable in pseudo-random number generation? [0.0]
  This paper explores the development and viability of an alternative pseudorandom number generator (PRNG) Traditional PRNGs, notably the one implemented in the Java.Random class, suffer from predictability which gives rise to exploitability. This study proposes a novel PRNG designed using ordinary differential equations, physics modeling, and chaos theory.
  arXiv  Detail & Related papers  (2024-04-15T00:28:51Z)
• Statistical testing of random number generators and their improvement using randomness extraction [0.0]
  Random number generators (RNGs) are notoriously challenging to build and test, especially for cryptographic applications.<n>We design, implement, and present various post-processing methods, using randomness extractors, to improve the RNG output quality.<n>We introduce a comprehensive statistical testing environment, based on existing test suites, that can be parametrised for lightweight (fast) to intensive testing.
  arXiv  Detail & Related papers  (2024-03-27T16:05:02Z)
• Reproducibility, energy efficiency and performance of pseudorandom number generators in machine learning: a comparative study of python, numpy, tensorflow, and pytorch implementations [0.0]
  Pseudo-Random Number Generators (PRNGs) have become ubiquitous in machine learning technologies because they are interesting for numerous methods. This study investigates whether the leading Pseudo-Random Number Generators (PRNGs) employed in machine learning languages, libraries, and frameworks uphold statistical quality and numerical when compared to the original C implementation.
  arXiv  Detail & Related papers  (2024-01-30T15:44:14Z)
• Simplex Random Features [53.97976744884616]
  We present Simplex Random Features (SimRFs), a new random feature (RF) mechanism for unbiased approximation of the softmax and Gaussian kernels. We prove that SimRFs provide the smallest possible mean square error (MSE) on unbiased estimates of these kernels. We show consistent gains provided by SimRFs in settings including pointwise kernel estimation, nonparametric classification and scalable Transformers.
  arXiv  Detail & Related papers  (2023-01-31T18:53:39Z)
• Testing randomness of series generated in Bell's experiment [62.997667081978825]
  We use a toy fiber optic based setup to generate binary series, and evaluate their level of randomness according to Ville principle. Series are tested with a battery of standard statistical indicators, Hurst, Kolmogorov complexity, minimum entropy, Takensarity dimension of embedding, and Augmented Dickey Fuller and Kwiatkowski Phillips Schmidt Shin to check station exponent. The level of randomness of series obtained by applying Toeplitz extractor to rejected series is found to be indistinguishable from the level of non-rejected raw ones.
  arXiv  Detail & Related papers  (2022-08-31T17:39:29Z)
• Instability, Computational Efficiency and Statistical Accuracy [101.32305022521024]
  We develop a framework that yields statistical accuracy based on interplay between the deterministic convergence rate of the algorithm at the population level, and its degree of (instability) when applied to an empirical object based on $n$ samples. We provide applications of our general results to several concrete classes of models, including Gaussian mixture estimation, non-linear regression models, and informative non-response models.
  arXiv  Detail & Related papers  (2020-05-22T22:30:52Z)
• On the Discrepancy between Density Estimation and Sequence Generation [92.70116082182076]
  log-likelihood is highly correlated with BLEU when we consider models within the same family. We observe no correlation between rankings of models across different families.
  arXiv  Detail & Related papers  (2020-02-17T20:13:35Z)
• Certified Robustness to Label-Flipping Attacks via Randomized Smoothing [105.91827623768724]
  Machine learning algorithms are susceptible to data poisoning attacks. We present a unifying view of randomized smoothing over arbitrary functions. We propose a new strategy for building classifiers that are pointwise-certifiably robust to general data poisoning attacks.
  arXiv  Detail & Related papers  (2020-02-07T21:28:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.