Abstract:Recently, Zhong et al. performed landmark Gaussian boson sampling experiments with up to 144 modes using threshold detectors. The authors claim to have achieved quantum computational advantage with the implementation of these experiments, named Jiuzhang 1.0 and Jiuzhang 2.0. Their experimental results are validated against several classical hypotheses and adversaries using tests such as the comparison of statistical correlations between modes, Bayesian hypothesis testing and the Heavy Output Generation (HOG) test. We propose an alternative classical hypothesis for the validation of these experiments using the probability distribution of mixtures of coherent states sent into a lossy interferometer; these input mixed states, which we term squashed states, have vacuum fluctuations in one quadrature and excess fluctuations in the other. We find that for configurations in the high photon number density regime, the comparison of statistical correlations does not tell apart the ground truth of the experiment (two-mode squeezed states sent into an interferometer) from our alternative hypothesis. The Bayesian test indicates that, for all configurations excepting Jiuzhang 1.0, the ground truth is a more likely explanation of the experimental data than our alternative hypothesis. A similar result is obtained for the HOG test: for all configurations of Jiuzhang 2.0, the test indicates that the experimental samples have higher ground truth probability than the samples obtained form our alternative distribution; for Jiuzhang 1.0 the test is inconclusive. Our results provide a new hypothesis that should be considered in the validation of future GBS experiments, and shed light into the need to identify proper metrics to verify quantum advantage in the context of GBS. They also indicate that a classical explanation of the Jiuzhang 1.0 experiment, lacking any quantum features, has not been ruled out.

What problem does this paper attempt to address?

The problem this paper attempts to address is whether the "Jiuzhang 1.0" and "Jiuzhang 2.0" Gaussian Boson Sampling (GBS) experiments truly demonstrate quantum computational advantage or if these experimental results can be explained through classical methods. Specifically, the authors propose an alternative classical hypothesis, which uses mixed coherent states (referred to as "squeezed states") as input states, generating probability distributions after passing through a lossy interferometer. This hypothesis is compared with the probability distributions of the two-mode squeezed states used in the experiments to evaluate whether the classical model can explain the experimental results. ### Main Research Questions: 1. **Verify Experimental Results**: The authors aim to verify whether the "Jiuzhang 1.0" and "Jiuzhang 2.0" experiments truly demonstrate quantum computational advantage or if they can be explained through classical methods. 2. **Propose Alternative Hypothesis**: The authors propose an alternative classical hypothesis based on mixed coherent states and evaluate its validity through various testing methods (such as statistical correlation comparison, Bayesian testing, and re-output generation testing). 3. **Evaluate Reliability of Testing Methods**: The authors explore the reliability and limitations of existing verification methods (such as statistical correlation comparison) in validating quantum computational advantage. ### Research Background: - **Quantum Computational Advantage**: Quantum computational advantage refers to the ability of current quantum devices to significantly outperform classical computers in specific computational tasks. - **Gaussian Boson Sampling (GBS)**: GBS is a quantum computational task achieved by inputting squeezed states into an interferometer and measuring the output modes. Theoretically, sampling from the detection mode probability distribution of an ideal experiment is a computationally hard problem. - **"Jiuzhang 1.0" and "Jiuzhang 2.0" Experiments**: These two experiments used interferometers with 100 modes and 144 modes, respectively, and claimed to achieve quantum computational advantage. The experimental results were verified through various testing methods. ### Main Findings: - **Low Photon Number Density Configuration**: For low photon number density configurations (e.g., ν=0.044), the squeezed state hypothesis is inconsistent with the experimental results, and the ground truth hypothesis of the experiment (i.e., using two-mode squeezed states) is a better explanation. - **High Photon Number Density Configuration**: For high photon number density configurations (e.g., certain configurations of "Jiuzhang 1.0" and "Jiuzhang 2.0"), the squeezed state hypothesis is statistically consistent with the experimental results, but Bayesian testing and re-output generation testing indicate that the ground truth hypothesis of the experiment is still the more likely explanation. - **Limitations of Verification Methods**: The authors point out that merely comparing statistical correlations is insufficient to reliably verify quantum computational advantage, and more verification methods and metrics need to be considered. ### Conclusion: - **Feasibility of Classical Hypothesis**: Although the squeezed state hypothesis can explain the experimental results in some cases, the ground truth hypothesis of the experiment remains the more likely explanation. - **Future Research Directions**: The authors suggest considering more classical hypotheses in future GBS experiment verifications and developing more reliable verification methods and metrics. In summary, this paper provides a detailed analysis of the "Jiuzhang 1.0" and "Jiuzhang 2.0" experiments, proposes a new classical hypothesis, and evaluates the reliability and limitations of existing verification methods, offering new perspectives and methods for future quantum computational advantage verification.

Classical models may be a better explanation of the Jiuzhang 1.0 Gaussian Boson Sampler than its targeted squeezed light model

Classical Modeling of a Lossy Gaussian Bosonic Sampler

Classical modelling of a lossy Gaussian bosonic sampler

Experimental demonstration of Gaussian boson sampling with displacement

Gaussian boson sampling validation via detector binning

Gaussian Boson Sampling with Pseudo-Photon-Number-Resolving Detectors and Quantum Computational Advantage

Simulating Gaussian boson sampling quantum computers

Gaussian Boson Sampling with Pseudo-Photon-Number Resolving Detectors and Quantum Computational Advantage

Classical algorithm for simulating experimental Gaussian boson sampling

Effect of partial distinguishability on quantum supremacy in Gaussian Boson sampling

Phase-Programmable Gaussian Boson Sampling Using Stimulated Squeezed Light

Validation tests of GBS quantum computers give evidence for quantum advantage with a decoherent target

Linear cross-entropy certification of quantum computational advantage in Gaussian Boson Sampling

Speedup in Classical Simulation of Gaussian Boson Sampling

The boundary for quantum advantage in Gaussian boson sampling

Solving Graph Problems Using Gaussian Boson Sampling

The Complexity of Bipartite Gaussian Boson Sampling

Transition of Anticoncentration in Gaussian Boson Sampling

Classical benchmarking of Gaussian Boson Sampling on the Titan supercomputer

A duality at the heart of Gaussian boson sampling