Quantum computing for plasma physics

The current paradigm of scientific computing relies on Moore’s law, which has enabled more sophisticated models, bigger problem sizes, and finer resolutions. However, even with a continuous growth of computing power, many multi-scale and multi-physics problems will remain inaccessible. It is often speculated that quantum computing may bring the desired game-changing capabilities. However, there is a wide gap between promised quantum advantages on fault-tolerant computers versus what is achievable on near-term devices. Our research bridges the gap between what is possible in principle and what is achievable in practice by developing quantum algorithms and realizing them on hardware. On the algorithm side, because quantum computers rely on unitary operations instead of logic gates, efficient quantum algorithms need to be designed differently than their classical counterparts. On the hardware side, different platforms have their unique limitations and advantages, which need to be balanced when compiling quantum programs for execution. Our research takes an algorithm-hardware co-design approach to ensure that quantum algorithms are grounded with hardware realities, while enabling applications that may shape the maturation of future quantum technologies.

Simulating three-wave interactions on quantum devices

Quantum devices are programmed to solve the nonlinear coupled-mode equations. The occupation probabilities of three states are plotted as functions of the number of time steps. Exact results are shown in orange, results obtained on commercial hardware with standard gates are shown in cyan, substantially improved results using customized gates are shown in blue, and the dashed black lines indicate the expected performance with decoherence. []