Pseudo Random Number Generation: a Reinforcement Learning approach

TL;DR

This paper explores using reinforcement learning to generate pseudo-random number generators from scratch, aiming to improve understanding and development of PRNGs through a novel machine learning approach.

Contribution

It introduces a reinforcement learning method for creating PRNGs, demonstrating feasibility and establishing a foundation for future research combining RL and PRNG design.

Findings

Reinforcement learning can be used to generate PRNGs from scratch.

The approach is feasible and comparable to classical methods.

It lays groundwork for further RL-based PRNG research.

Abstract

Pseudo-Random Numbers Generators (PRNGs) are algorithms produced to generate long sequences of statistically uncorrelated numbers, i.e. Pseudo-Random Numbers (PRNs). These numbers are widely employed in mid-level cryptography and in software applications. Test suites are used to evaluate PRNGs quality by checking statistical properties of the generated sequences. Machine learning techniques are often used to break these generators, for instance approximating a certain generator or a certain sequence using a neural network. But what about using machine learning to generate PRNs generators? This paper proposes a Reinforcement Learning (RL) approach to the task of generating PRNGs from scratch by learning a policy to solve an N-dimensional navigation problem. In this context, N is the length of the period of the generated sequence, and the policy is iteratively improved using the average value of an appropriate test suite run over that period. Aim of this work is to demonstrate the feasibility of the proposed approach, to compare it with classical methods, and to lay the foundation of a research path which combines RL and PRNGs.