Home Page News

Progress in reliable quantum computing

Recently, Microsoft announced that they have achieved promising results by applying their qubit-virtualization system to Atom Computing’s neutral-atom qubits. The teams created 24 logical qubits and entangled them in a complex arrangement known as a cat state, or Greenberger-Horne-Zeilinger (GHZ) state. They also detected errors, corrected those errors, and performed computation with 28 logical qubits. These achievements demonstrate the rapid advances being made in reliable quantum computing as the industry graduates from the noisy, intermediate-scale quantum computing era.

Applying Microsoft’s system to Atom Computing’s qubits

To store and process quantum information, Atom Computing uses neutral atoms, which have several advantages over other types of qubits. Neutral atoms offer extensive scalability because they can be tightly packed in a small amount of space. Furthermore, neutral atoms have low susceptibility to environmental noise and they can be moved around to allow interaction with other atoms in the array. Notably, Atom Computing recently achieved 99.6% two-qubit gate fidelity, demonstrating that their neutral-atom qubits have error rates low enough to enable error detection and correction. For all these reasons, neutral-atom qubits offer a promising pathway to scalable and reliable quantum computing.

The Azure Quantum compute platform was developed by Microsoft to enable reliable quantum computation across a variety of qubit types. This platform combines Microsoft’s qubit-virtualization system with quantum-processing units (QPUs) from hardware partners such as Atom Computing. By detecting and correcting the errors in the physical qubits of the QPUs, this system creates the logical qubits needed for reliable quantum computing.

Turning physical qubits into reliable logical qubits

When applying Microsoft’s qubit-virtualization system to Atom Computing’s neutral-atom qubits, the teams took steps to not only detect errors but also to detect and correct losses of neutral atoms during experiments. This is necessary because some of the atoms being held in place with lasers become lost, resulting in a loss of their quantum information.

In one experiment, the teams created 24 logical qubits and entangled them in a cat (GHZ) state. The error rate of the entangled logical qubits varied depending on whether losses were only detected or were detected and corrected. In both cases, the logical qubits demonstrated a lower error rate than that of the corresponding physical qubits.

In a separate experiment, 28 logical qubits were used to perform computation while detecting and correcting errors, a combination that is critical to achieving scientific quantum advantage. This computation—based on the Bernstein-Vazirani algorithm—was able to be performed more accurately with the logical qubits than with the corresponding physical qubits.

Making quantum capabilities commercially available

To bring their results out of the laboratory, Microsoft and Atom Computing are offering a quantum discovery suite with state-of-the-art capabilities created by applying Microsoft’s qubit-virtualization system to Atom’s neutral-atom hardware. By integrating Azure Elements, the offering combines logical qubits, cloud high-performance computing, and advanced AI models that, together, enable exploration into uncharted territories across multiple fields, including chemistry and materials science.

SPONSORED ARTICLE; author Deborah A. Hutchinson