Regimes of Classical Simulability for Noisy Gaussian Boson Sampling

Area of Science:

• Quantum Information Science
• Quantum Computing
• Quantum Optics

Background:

• Gaussian boson sampling (GBS) is a leading candidate for demonstrating quantum computational supremacy.
• Experimental noise poses a significant challenge, potentially making GBS classically simulable.
• Existing GBS architectures often face increasing photon loss with circuit depth.

Purpose of the Study:

• To formalize the conditions under which noisy GBS can be classically simulated.
• To establish a nonclassicality test for quantum supremacy claims based on GBS.
• To identify operational regimes for GBS devices that maintain a quantum advantage.

Main Methods:

• Derivation of a sufficient condition for approximate polynomial-time classical simulation of noisy GBS.
• Analysis of an inequality involving input squeezing, transmission rate, and detector efficiency.
• Investigation of noise impact on GBS performance across different linear-optical architectures.

Main Results:

• A condition is established linking input squeezing, transmission, and detector quality to classical simulability.
• Noisy GBS generally loses its quantum advantage in the asymptotic limit for typical architectures.
• Intermediate-sized GBS devices may offer a quantum advantage under modest noise levels.

Conclusions:

• The derived condition acts as a crucial test for GBS-based quantum supremacy.
• Photon loss significantly impacts the scalability of GBS quantum advantage.
• Increasing input squeezing shows promise for mitigating noise and preserving quantum advantage.