Increasing precision of uniform pseudorandom number generators
This addresses a technical challenge for developers and researchers using GPUs in simulations or applications requiring high-precision random numbers, but it appears incremental as it builds on existing pseudorandom number generation techniques.
The paper tackles the problem of generating uniformly distributed pseudorandom numbers with extended precision by proposing a method that combines two lower-precision numbers, specifically enabling efficient generation on GPUs where performance varies by precision.
A general method to produce uniformly distributed pseudorandom numbers with extended precision by combining two pseudorandom numbers with lower precision is proposed. In particular, this method can be used for pseudorandom number generation with extended precision on graphics processing units (GPU), where the performance of single and double precision operations can vary significantly.