Curator's Take
This article represents a significant milestone in demonstrating quantum utility for real scientific problems, using IBM's 63-qubit heavy-hex architecture to tackle Ising model ground states - a fundamental challenge in condensed matter physics and optimization. The researchers' success in finding the boundary between acceptable and unacceptable quantum error rates provides crucial guidance for when current noisy quantum computers can outperform classical methods. Particularly noteworthy is their discovery that Ising models appear especially well-suited for near-term quantum devices, potentially opening the door for practical quantum advantage in studying magnetic materials, optimization problems, and phase transitions before we achieve full fault tolerance.
— Mark Eatherly
Summary
We find the ground-state energy of the Ising model using the Cascaded Variational Quantum Eigensolver (CVQE) algorithm with the Guided-Sampling Ansatz (GSA) using up to 63 qubits on a quantum computer. We study a heavy-hex lattice to match the qubit architecture, allowing us to perform calculations in the quantum utility regime. We study both a homogeneous and random-coupling model. We locate the boundary of acceptable quantum errors as a function of the number of qubits and coupling strength. An entropic analysis is performed giving insights into the quantum computing performance. A subspace analysis is performed that suggests that the Ising model is especially suited for near-term quantum computing.