
CSIRO research has found that an AI neural network syndrome decoder can detect errors in quantum processors and make appropriate corrections. Credit: CSIRO
Could artificial intelligence help overcome one of quantum computing's biggest obstacles?
New research from Australia's national science agency has found that AI could help solve quantum computing errors, a key step towards quantum computers one day solving complex real-world problems.
The CSIRO study: Physical Review Research The magazine found for the first time that AI can help handle and resolve quantum errors called qubit noise, which are generated by the nature of quantum physics.
Overcoming these errors is widely considered to be the biggest barrier to moving advanced quantum computers from experiments to tools.
In classical computers, information is stored and processed in “bits” that operate on the principle of binary numbers: each bit represents either a 0 or a 1. However, quantum computing devices are made up of quantum bits, or “qubits.”
They exploit the special properties of quantum mechanics, allowing them to represent 0, 1, or both 0 and 1 simultaneously, which promises to unlock enormous computing power and make it possible to solve problems that are insurmountable by classical computers.
However, due to the delicate nature of qubits, quantum computers can also generate “noise”, or errors, in their output. To overcome this, quantum error-correcting codes are used to detect and correct errors.
CSIRO has implemented an AI neural network syndrome decoder to detect errors and make appropriate corrections, and Dr Muhammad Usman, CSIRO's Data61 quantum systems team leader, says this work will enable efficient handling of complex errors from real quantum hardware.
“Our work demonstrates for the first time that a machine learning-based decoder is in principle able to process error information derived directly from IBM device measurements and suggest appropriate corrections despite the highly complex nature of the noise,” he said.
“In our study, we did not observe error suppression with increasing distance for error-correcting codes, as theoretically predicted, due to the current large noise levels (above the code threshold) in the IBM quantum processor.”
Quantum error correcting codes were developed to combat the fundamental physical noise of qubits by distributing logical information across many physical qubits.
These codes interpret the error information by measuring stabilizers in a lattice of qubits, called syndrome measurements. An efficient, fast, and scalable implementation of the computationally expensive syndrome processing step is critical to the overall performance of quantum error-correcting codes.
To improve this correction efficiency, Dr. Usman implemented and trained an artificial neural network syndrome decoder.
The performance of the neural network decoder was directly benchmarked on an IBM quantum processor, demonstrating that it can efficiently handle complex errors from real quantum hardware and make appropriate corrections.
Research suggests that as physical error rates decrease in the next few years, AI will be able to achieve error suppression as code distances increase, and potentially even full fault tolerance once code distances are large enough.
For more information:
Brhyeton Hall et al., Decoding Artificial Neural Network Syndrome with IBM Quantum Processors, Physical Review Research (2024). DOI: 10.1103/PhysRevResearch.6.L032004
Quote: Artificial intelligence could help make quantum computing a reality (July 12, 2024) Retrieved July 12, 2024, from https://phys.org/news/2024-07-artificial-intelligence-quantum-reality.html
This document is subject to copyright. It may not be reproduced without written permission, except for fair dealing for the purposes of personal study or research. The content is provided for informational purposes only.