Quemix and Sumitomo Rubber Industries propose Fourier-space readout method to reconstruct functions from quantum states
Quemix and Sumitomo Rubber Industries proposed a Fourier-space readout (FSR) method to efficiently reconstruct continuous functions encoded in quantum states. They showed by theoretical analysis and numerical experiments that this approach can suppress the readout cost—a bottleneck for applying quantum computing to CAE and similar fields—while maintaining computational speedups.
Summary of the announcement
The research was published in the journal Quantum Science and Technology. In the FSR method, the dominant Fourier coefficients of a continuous function are obtained on a quantum computer, and that information is used to reconstruct the function on a classical computer. Analysis showed that the quantum-computing-side cost grows logarithmically with the number of grid points, while the classical-computing-side cost increases with the number of points to be reconstructed rather than with the total number of grid points.
Key points
- Quemix and Sumitomo Rubber Industries conducted joint research addressing the readout problem of quantum-computed results for CAE and similar applications.
- They proposed a quantum–classical hybrid method that obtains dominant Fourier coefficients from a quantum state and reconstructs continuous functions on a classical computer.
- Theoretical analysis and numerical experiments showed that the quantum-computing-side cost increases logarithmically with the number of grid points.
- The classical-computing-side cost was found to increase according to the number of points to be reconstructed, not the total number of grid points.
Technical and business significance
Even if solutions used in CAE can be computed efficiently as quantum states, the advantage of quantum computing could be lost if reading out information at all grid points incurs large costs. The FSR method is technically significant in that it proposes to alleviate this bottleneck by restricting processing to dominant Fourier coefficients and to the points required for reconstruction. However, the performance on real hardware, concrete speedup factors, and steps toward commercialization were not disclosed in this announcement.
Points to watch going forward
Going forward, it will be important to evaluate performance on actual quantum hardware and whether an overall advantage over classical methods—including readout—can be achieved. Results applying the method to larger-scale CAE problems than those used in the numerical experiments, and the range of functions for which a small number of Fourier coefficients yield sufficient accuracy, will also be important metrics. In addition, whether the collaboration with Sumitomo Rubber Industries advances to validation and deployment in concrete CAE workflows will be worth watching.
✍️ Quantum Index Analysis
How well the target boundary conditions can be reproduced is difficult for non-specialists to judge. In pure Fourier spectral methods, boundary conditions assume very simple forms—such as periodic boundaries—and therefore cannot handle practical CAE. For the same reason, this method is likely to target fundamental numerical analysis rather than practical industrial use, even in a future in which FTQC is realized.
