IBM and researchers from the University of Chicago have demonstrated a quantum computation that they said exceeded the practical capabilities of leading classical simulation methods while providing statistical evidence that the result was accurate. The researchers used a new error correction method to encode 70 logical qubits and complete the calculation in approximately 15 minutes.
IBM said leading classical computing approaches would have required an infeasible amount of time to perform the same task.
The research, detailed in a paper titled “Sampling Hard Circuits With Verifiably High Fidelity,” used a structured alternative to random circuit sampling.
Random circuit sampling has been used to test whether quantum systems can outperform classical computers, but verifying the results becomes increasingly difficult as computational complexity grows.
The IBM and University of Chicago approach retained the computational difficulty associated with random circuit sampling while introducing a structure that allowed researchers to detect errors.
The team executed 2,415 logical two-qubit operations and 468 logical T gates.
Encoding the circuit reduced effective logical error rates to approximately one-tenth of the physical error rates, supporting high fidelity despite the large number of operations.
The circuits and experimental results have been released through the Quantum Advantage Tracker.
KEY QUOTES:
“This experiment develops techniques to better characterize the fidelity of hard quantum states under noise, increasing confidence that the quantum computer is solving a computationally hard problem.”
Bill Fefferman, Associate Professor at the University of Chicago
“We are now firmly in the quantum advantage era.”
“This milestone gives scientists, developers and businesses a new foundation for trusting quantum computers as they scale to problems far beyond what we can achieve classically.”
Jay Gambetta, Director of IBM Research and IBM Fellow
