Improving Random Number Generation with a Dual Neural Learning Algorithm
Author Information
Author(s): Simone Fiori
Primary Institution: Università Politecnica delle Marche
Hypothesis
Can a dual neural learning algorithm improve the generation of asymmetric random numbers?
Conclusion
The new dual neural learning algorithm effectively generates random numbers with desired asymmetric distributions without the need for normalization.
Supporting Evidence
- The proposed method simplifies the learning problem by focusing on the dual cardinal equation.
- The algorithm demonstrated high flexibility and efficiency in generating random samples.
- Numerical experiments confirmed the effectiveness of the proposed approach across various distributions.
Takeaway
This study shows a new way to create random numbers that can take on different shapes, making it easier to use them in things like games and experiments.
Methodology
The study used a dual neural system to learn the inverse function for generating random samples from specified distributions.
Limitations
The method is limited to generating distributions that are either symmetric or skewed to the right, although the new approach aims to extend this.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website