PyQBench
Qubit measurements on noisy devices are certified by PyQBench, an open-source Python tool.
An important update to PyQBench, an open-source Python framework for evaluating gate-based quantum computers, has been revealed by researchers. A significant addition to the latest edition is the certification of qubit von Neumann measurements. In order to assess the performance of Noisy Intermediate-Scale Quantum (NISQ) devices, this addition is essential.
The dependability of quantum computing is hampered by the large noise, mistakes, and small qubit counts of NISQ devices. To evaluate their performance, strong benchmarking techniques are becoming more and more necessary. Existing methods such as quantum volume metrics, randomised benchmarking, and cross-entropy benchmarking offer insightful information, but they frequently concentrate on device properties or gate-level defects. Precise evaluation of qubit measurements is especially crucial since imprecise measurements cause mistakes to build up and limit device capability.
This is addressed by the updated PyQBench, which assesses accuracy on NISQ devices by implementing a certification method for qubit von Neumann measurements. The original functionality of PyQBench, which concentrated on the discrimination of von Neumann measurements, is expanded in this way. Researchers may more precisely evaluate the quality of quantum measurements to the certification method.
You can also read Karnataka Funds ₹48 Crore for Quantum Research Park phase 2
A hypothesis-testing methodology is used in the core certification process. It seeks to ascertain if a device is carrying out a particular measurement (the null hypothesis, such as the measurement in the computational Z-basis) or a different one. In order to accomplish this, an initial state must be prepared, the first qubit must be measured, and then, depending on the result of the first, a second, auxiliary qubit must be measured.
The null hypothesis is either accepted or rejected based on the results of the final measurement. For a certain probability of type I mistake (rejecting the null hypothesis when it is true), the goal is to minimise the chance of a type II error (accepting the null hypothesis when the alternative is true).
The incapacity of current NISQ devices to conduct conditional measurements makes a direct application of this technique difficult. Postselection and direct sum are two comparable methods that PyQBench uses to get around this. Both approaches adapt the plan to make use of the hardware components that are available.
By integrating Python libraries and providing an intuitive command-line interface (CLI), PyQBench provides versatility. Users can construct circuits, integrate error models, specify custom measurement schemes, perform benchmarks on designated backends, and evaluate results with the Python library.
Especially for the parametrised Fourier family of qubit von Neumann measurements, the CLI offers a streamlined method for executing preset benchmarking scenarios. Running synchronous or asynchronous benchmarks, monitoring the status of asynchronous processes, resolving work outcomes, and tabulating data are all supported using the CLI.
By integrating with the Qiskit library, the latest version is specifically made to handle IBM Q devices, allowing for smooth benchmarking on actual quantum hardware.
You can also read ColibriTD Launches QUICK-PDE Hybrid Solver On IBM Qiskit
Results from benchmarking on IBM Q devices show how successful the certification program is. Empirical results closely match theoretical expectations in experiments assessing the probability of type II error for the parametrised Fourier family of observations. Readout errors, decoherence, and quantum noise all cause deviations.
Importantly, the accuracy of the experimental data was much increased by the use of error mitigation measures. The Mthree mitigation strategy effectively decreased experimental and theoretical value deviations, particularly in situations where type II error probability were low. This emphasises how crucial it is to incorporate noise-aware techniques. In areas where the ideal probability is near zero and uncontrolled data exhibit greater deviations, error mitigation regularly improves the results. The framework’s resilience was strengthened by the mean absolute error, which was continuously less than 0.01.
PyQBench meets the crucial demand for trustworthy performance measures in the NISQ era by offering adaptable and trustworthy benchmarking tools. It aids in the creation of methods for reducing errors and confirming the accuracy of quantum measurements.
Future development might include adding more advanced error mitigation strategies and benchmarking measures, as well as extending PyQBench’s functionality to other quantum hardware platforms.
PyQBench’s code is open-source and accessible on GitHub, promoting community cooperation and additional developments in quantum hardware benchmarking.
You can also read Government Agencies Get QuProtect QuSecure’s via Carahsoft
