Enabling Quantum Computer Simulation under Minimal Precision Floating-Point using Irrational Value Decomposition

Quantum computing holds immense promise for solving complex problems beyond the reach of classical computers. However, the realization of practical quantum computers faces significant challenges, including its inherent noise and error rates. To address these challenges and explore the capabilities of quantum computers, accurate and efficient quantum computer simulations are essential. This paper presents a novel quantum computer simulation framework designed to accommodate a wide range of custom reduced precision floating-point formats. Leveraging our proposed quantum computation algorithm, our framework enables the quantum computer simulations using exceptionally low precision floating-point numbers, while maintaining high accuracy levels. Through a systematic analysis, we explore the impact of floating-point precision on the accuracy of the quantum computer simulation, providing optimized bit-widths tailored for floating-point numbers. Particularly, using our proposed quantum computation algorithm, our framework achieves quantum state vector outcomes comparable to full-precision quantum simulations, even with extremely low precision floating-point formats ranging from 4 to 6 bits. Overall, these findings highlight the viability of precision reduction techniques in enhancing the efficiency of quantum computer simulations.