Quantum computational advantage via high-dimensional Gaussian boson sampling.

Area of Science:

• Quantum Information Science
• Photonic Quantum Computing
• Computational Complexity Theory

Background:

• Photonics offers a promising route to quantum computational advantage (QCA), yet current Gaussian boson sampling (GBS) implementations face experimental challenges like loss and lack of programmability.
• Rigorous theoretical evidence for the classical hardness of GBS, a key requirement for demonstrating QCA, remains comparatively underdeveloped.

Purpose of the Study:

• To strengthen the theoretical evidence for the classical hardness of GBS, comparable to leading QCA proposals.
• To introduce a novel, programmable QCA architecture called high-dimensional GBS (HD-GBS) with improved experimental feasibility.
• To demonstrate that HD-GBS experiments can outperform classical simulation algorithms for GBS at modest system sizes.

Main Methods:

• Developed theoretical frameworks to provide rigorous evidence for the computational hardness of GBS.
• Proposed the high-dimensional GBS architecture, designed for programmability and low-loss implementation using minimal optical components.
• Analyzed and compared the performance of classical GBS simulation algorithms against proposed HD-GBS experiments.

Main Results:

• Provided theoretical evidence for GBS hardness comparable to the strongest QCA proposals.
• Introduced HD-GBS, a programmable photonic architecture implementable with low loss and few components.
• Showcased that HD-GBS experiments outperform classical simulation algorithms for GBS at modest scales.

Conclusions:

• The study establishes a stronger theoretical foundation for GBS as a platform for QCA.
• The proposed HD-GBS architecture offers a practical and programmable approach to photonic quantum computation.
• This work paves the way for demonstrating QCA using programmable photonic processors.