Conferences
29 September 2025, Online
Mathematics Webinar | Benford’s Law in the Age of AI: New Frontiers in Data Authenticity and Model Transparency, 29 September 2025
As artificial intelligence (AI) systems increasingly shape critical decisions in finance, auditing, cybersecurity, and governance, the demand for interpretable, transparent, and data-authentic models continues to grow. In this context, Benford’s Law—a mathematical law that predicts the frequency distribution of leading digits in naturally occurring datasets—emerges as a valuable tool for anomaly detection, model validation, and forensic analysis.
This webinar explores the evolving role of Benford’s Law in the age of AI, emphasizing how it can be integrated into machine learning workflows as a statistical feature for identifying irregularities and improving model accountability. By embedding digit-based conformity tests into AI pipelines, analysts and developers can enhance the explainability of models and proactively detect data manipulation, bias, or fraud in large-scale systems.
One of the focuses of the session will be the statistical analysis of how Pareto and Weibull distributions, frequently used in economics, risk modeling, and reliability engineering, align with Benford’s expected digit frequencies. We will discuss both the theoretical underpinnings and present empirical evidence to evaluate the extent to which these distributions comply with Benford’s Law and what that means for AI systems trained on such data.
Participants will gain insights into cutting-edge applications, including hybrid AI models that combine traditional statistical methods with deep learning, as well as the use of Benford-based metrics in improving robustness, reducing false positives, and enhancing the interpretability of black-box models.
Whether you are a researcher, auditor, data scientist, or AI practitioner, this webinar will offer a multidisciplinary perspective on how mathematical laws, statistical rigor, and artificial intelligence can work together to build more trustworthy and resilient systems.
Date: 29 September 2025
Time: 12:00 p.m. CEST | 06:00 a.m. EDT
Webinar ID: 865 9326 5734
Website: https://sciforum.net/event/Mathematics-20?subscribe
Register now for free!
|
Speaker/Presentation |
Time in EDT |
Time in CEST |
|
Vesna Rajić |
06:00–11:10 |
12:00–12:10 |
|
Dragan Azdejković |
06:10–06:40 |
12:10–12:40 |
|
Jelena Stanojević & Tatjana Rakonjac-Antić |
06:40–07:20 |
12:40–13:20 |
|
Dragana Radojičić |
07:20–07:40 |
13:20–13:40 |
|
Q&A Session |
07:40–08:00 |
13:40–14:00 |
|
Vesna Rajić |
08:00–08:00 |
14:00–14:00 |
After registering, you will receive a confirmation email outlining how to join this webinar. Registrations with academic institutional email addresses will be prioritized.
Unable to attend? Register anyway, and we will let you know when the recording is available to watch.
Webinar Chairs and Keynote Speakers:
- Prof. Dr. Vesna Rajić, Department for Statistics and Mathematics, Faculty of Economics and Business, University of Belgrade, Belgrade, Serbia;
- Dr. Dragan Azdejković, Department of Statistics and Mathematics, University of Belgrade, Faculty of Economics and Business, 11000 Belgrade, Serbia;
- Dr. Jelena Stanojević, Department of Statistics and Mathematics, University of Belgrade, Faculty of Economics and Business, 11000 Belgrade, Serbia;
- Prof. Dr. Tatjana Rakonjac-Antić, Department of Economic Policy and Development, University of Belgrade, Faculty of Economics and Business, Belgrade, Serbia;
- Dr. Dragana Radojičić, Department of Statistics and Mathematics, University of Belgrade, Faculty of Economics and Business, 11000 Belgrade, Serbia.
Relevant Special Issue:
“Statistics and Nonlinear Analysis: Simulation and Computation”
Guest Editor: Prof. Dr. Vesna Rajić
Deadline for manuscript submissions: 30 September 2025
