A quantum computing method combines error correction and result checking to perform calculations while improving accuracy and reliability.
IBM and researchers from the University of Chicago have demonstrated a quantum computing method that can perform complex calculations beyond practical classical simulation while also checking whether the results are accurate.
The team used a new approach called encoded quantum circuits, described in the research paper “Sampling hard circuits with verifiably high fidelity.” The method combines quantum error correction with result verification, addressing two major challenges in achieving practical quantum advantage.
The demonstration used 70 logical qubits, which are groups of physical qubits designed to protect quantum information from errors. The system completed 2,415 logical two-qubit operations and 468 logical “T gates,” operations commonly used to measure the complexity of quantum circuits.
The researchers showed that the encoded circuit could reduce the effective error rate of the computation to about one-tenth of the physical error rate. This allowed the quantum system to maintain accuracy while running a large number of operations.
Quantum advantage experiments often use a technique called random circuit sampling (RCS), where a quantum computer generates output patterns that are difficult for classical computers to reproduce. However, confirming whether those outputs are correct has been a major challenge, as classical verification becomes harder as circuits grow more complex.
The new method modifies RCS by adding a structure that helps detect errors during computation. This allows researchers to verify quantum results without requiring a complete classical simulation of the entire calculation.
During testing, the team found that several advanced classical simulation methods would require impractical amounts of time to complete the same task. The quantum system completed the calculation in around 15 minutes.
Soumik Ghosh, PhD student in Fefferman’s group at the University of Chicago, added, “Beyond strengthening experimental validation, advances in verification have the potential to unlock practical applications for the next generation of quantum computers.”
The researchers said that combining error correction with reliable verification is an important step toward building larger quantum computers that can deliver useful results.
