What Cross-Entropy Benchmarking Is
Cross-entropy benchmarking (XEB) is a method designed to verify that a quantum computer can perform tasks beyond the capabilities of classical computers. It measures how well a quantum device generates samples from a specific probability distribution, often referred to as the Porter-Thomas distribution.
The technique involves comparing the output of an honest noisy quantum sampler with those generated by a classical spoofing adversary. The goal is to show that the quantum device's results are statistically distinguishable from classical simulations.
Why It Matters
XEB is crucial for demonstrating quantum supremacy, which means a quantum computer can perform tasks that would be infeasible for any classical computer. This not only advances the field of quantum computing but also has implications for cryptography, optimization problems, and other areas where quantum computers could offer significant advantages.
Moreover, XEB helps researchers understand the limitations and noise levels in current quantum devices, guiding improvements towards more robust and reliable quantum technologies.
How It Works
The process begins with a quantum device generating bitstrings according to its output distribution. These bitstrings are then compared against the expected Porter-Thomas distribution, which is derived from the theory of random unitary matrices. The cross-entropy benchmarking statistic quantifies how close the quantum device's results are to this ideal distribution.
On the other hand, a classical adversary attempts to mimic the quantum output using its own algorithms and resources. If the honest sampler’s results significantly outperform those of the classical spoofing adversary, it indicates that the quantum device is performing genuinely quantum computations.
Real-World Applications
Cross-entropy benchmarking has been used to verify Google's Sycamore quantum processor’s supremacy over classical supercomputers. This achievement not only marks a milestone in the development of quantum technology but also opens up new avenues for research and practical applications, such as simulating complex molecular structures or optimizing financial portfolios.
Future advancements in XEB could lead to more efficient verification methods for larger and noisier quantum devices, further pushing the boundaries of what is possible with quantum computing.
Frequently asked questions
How does cross-entropy benchmarking differ from other quantum supremacy tests?
Cross-entropy benchmarking specifically measures how well a quantum device can generate samples that match the expected Porter-Thomas distribution, distinguishing it from other methods like randomized benchmarking or tomography which focus on different aspects of quantum performance.
What is the significance of the Porter-Thomas distribution in XEB?
The Porter-Thomas distribution is significant because it represents the theoretical output distribution expected for a random unitary matrix, serving as a benchmark to compare against the actual results from a quantum device. Deviations from this distribution can indicate genuine quantum behavior.
Can classical computers perform cross-entropy benchmarking?
Yes, classical computers can perform cross-entropy benchmarking by generating bitstrings that mimic the expected Porter-Thomas distribution using sophisticated algorithms and resources. However, they cannot match the raw computational power of a quantum device.
What challenges does XEB face in practical applications?
XEB faces challenges such as noise in quantum devices, which can distort the output distributions, making it harder to distinguish between genuine quantum behavior and classical simulations. Additionally, the complexity of implementing and interpreting the benchmarking results requires careful calibration and analysis.
Try it live
Everything above runs in your browser — open Cross-Entropy Benchmarking Verifier and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.
▶ Open Cross-Entropy Benchmarking Verifier simulation