A new technology by researchers at Princeton University, the University of Chicago and IBM significantly improves the reliability of quantum computers by exploiting noise data in operations on real hardware. In a paper presented this week, researchers describe a new compilation method that increases the possibility of resource-limited and "noisy" quantum computers to produce useful answers. In particular, the researchers showed an almost three-fold average improvement in the reliability of real-system runs on IBM's 16-bit quantum computer, which improved some program executions by as much as eighteen times.
The joint research group includes data scientists and physicists from EPiQC (Enabling Practical-scale Quantum Computation), an NSF Expedition in Computing that started in 2018. The EPiQC aims to bridge the gap between theoretical quantum uses and programs for practical quantum computer architectures on nearest units. EPiQC researchers collaborated with quantum computing experts from IBM for this study, which will be presented at the 24th ACM International Programming Language and Operating System (ASPLOS) Conference Conference on Providence, Rhode Island, April 17.
Customize programs to qubit noise
Quantum computers are composed of qubits (quantum bits) that are equipped with special properties from quantum mechanics. These special features (superposition and entanglement) allow the quantum computer to represent a very large space of opportunity and comb through them for the right answer, finding solutions much faster than classic computers.
However, the quantum computers today and the next 5-10 years are limited by noisy operations, where the quantum computer port gives rise to errors and errors. During the execution of a program, these errors accumulate and may lead to incorrect answers.
To compensate for these errors, users run quantity programs thousands of times and choose the most common answer as the right answer. The frequency of this response is called the program's success rate. In an ideal quantum computer, this success rate would be 100% – every time the hardware would produce the same answer. In practice, however, the success rates are much less than 100% due to noisy activity.
The researchers observed that the mismanagement of quantum operations on real hardware, such as the IBM 16-bit system, has very large variations between different hardware solutions (qubits / gates) in the system. These error rates can also vary from day to day. The researchers found that malfunction can have up to 9 times as much variation depending on the time and place of the operation. When a program runs on this machine, the hardware quota selected for the run determines the success rate.
"If we want to run a program today, and our compiler chooses a hard drive port (operation) that has poor rates of interest, the program's success rate decreases dramatically," says researcher Prakash Murali, a doctoral student at Princeton University. "If we compile with awareness of this sound and running our programs with the best qubits and operations in the hardware, we can significantly increase the success level. "
To take advantage of this idea of adapting program execution to hardware noise, the researchers developed a "sound-adaptive" compiler that utilizes detailed noise qualification information for the target hardware. Such audio data is routinely measured for IBM quantum systems as part of the daily operating calibration and includes the error rates for each type of operation capable of the hardware. Utilizing this data, the compiler maps program quota to hardware qubits that have low error rates and schedules ports quickly to reduce the risk of state decay from decoherence. In addition, it also minimizes the number of communication operations and performs them with reliable hardware operation.
Improve the quality of runs on a real quantum system
To show the effect of this approach, the researchers gathered and implemented a set of benchmark programs on 16 quartz IBM quantum computers and compared the success of their new audio adaptive compiler to executions from IBM's Qiskit compiler, the standard compiler for that machine. Over benchmarks, they observed nearly a three-fold average improvement in success rate, with up to eighteen times improvements on some programs. In many cases, IBM's compiler gave incorrect answers to the executions due to its noiselessness, while the audio adaptive compiler provided correct answers with high success rates.
Although the team's methods were shown on the 16-bit machine, all quan- tities over the next 5-10 years are expected to have noisy operations due to difficulties in performing precise gates, defects caused by lithographic manufacturing, temperature fluctuations, and other sources. Noise adaptivity will be crucial for utilizing the computational power of these systems and paving the way for large-scale quantum calculations.
"When we run large programs, we want the success rate to be high in order to be able to distinguish the right response from noise and also reduce the number of repeated runs required to get the answer," stressed Murali. "Our evaluation clearly shows that sound adaptivity is crucial to achieving the full potential of quantum systems."
The team's entire paper, "Noise-Adaptive Compiler Mappings for Noisy Intermediate-Scale Quantum Computers", is now published on arXiv and will be presented at the conference 24th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) in Providence, Rhode Island April 17th.
Research provides speed increase to quantum computers
Prakash Murali et al. Noise-Adaptive Compiler Mappings for Noisy Intermediate-Scale Quantity Computers, arXiv: 1901.11054 [quant-ph] arxiv.org/abs/1901.11054
Researchers use audio data to increase the reliability of quantum computers (2019, April 17)
downloaded April 17, 2019
This document is subject to copyright. Except for any fair trade for private study or research, no
Some may be reproduced without written permission. The content is provided for informational purposes only.