Efficient algorithm for full-state quantum circuit simulation with DD compression while maintaining accuracy

With the development of noisy intermediate-scale quantum machines, quantum processors show their supremacy in specific applications. To better understand the quantum behavior and verify larger quantum bit (qubit) algorithms, simulation on classical computers becomes crucial. However, as the simulate...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Quantum information processing Jg. 22; H. 11
Hauptverfasser: Song, Yuhong, Sha, Edwin Hsing-Mean, Zhuge, Qingfeng, Xu, Rui, Wang, Han
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Springer US 17.11.2023
Schlagworte:
Data Structures and Information Theory
Mathematical Physics
Physics
Physics and Astronomy
Quantum Computing
Quantum Information Technology
Quantum Physics
Spintronics
Quantum circuit simulation
Decision diagram
Algorithm optimization
Error mitigation
ISSN:1573-1332, 1573-1332
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:With the development of noisy intermediate-scale quantum machines, quantum processors show their supremacy in specific applications. To better understand the quantum behavior and verify larger quantum bit (qubit) algorithms, simulation on classical computers becomes crucial. However, as the simulated number of qubits increases, the full-state simulation suffers exponential memory increment for state vector storing. In order to compress the state vector, some existing works reduce the memory by data encoding compressors. Nevertheless, the memory requirement remains massive. Meanwhile, others utilize compact decision diagrams (DD) to represent the state vector, which only demands linear memory size. However, the existing DD-based simulation algorithm possesses many redundant calculations that require further exploration. Besides, the traditional normalization-based nodes merging method of DD amplifies the side influences of approximate error. Therefore, to tackle the above challenges, in this paper, we first fully explore the redundancies in the recursive-based DD simulation (RecurSim) algorithm. Inspired by the regularities of the quantum circuit model, a scale-based simulation (ScaleSim) algorithm is proposed, which removes plenty of unnecessary computations. Furthermore, to eliminate the influences of approximate error, we propose a new pre-check DD building method, namely PCB, which maintains the accuracy of DD representation and produces more memory saving. Comprehensive experiments show that our method achieves up to 24124.2 × acceleration and 3.2  × 10 7 × memory reduction than traditional DD-based methods on quantum algorithms while maintaining the representation accuracy.
ISSN:1573-1332
1573-1332
DOI:10.1007/s11128-023-04160-5