D-Wave claims quantum supremacy with breakthrough in solving complex simulations

Canadian quantum computing company D-Wave has solved complex scientific simulations that classical supercomputers could not resolve using its 'quantum annealing' processor. Quantum supremacy, where quantum computers outperform classical computers, is increasingly becoming a reality.

An international team of researchers led by D-Wave announced the results of a simulation of a problem known as the 'transverse-field Ising model (TFIM)' using its quantum annealing processor in the international journal Science on the 13th. The research team released the results on the preprint server arXiv in March last year and underwent review for about a year before publishing the formal paper.

The TFIM solved by the research team is a model that studies what phenomena occur when a system of arranged magnets (spins) is subjected to an external magnetic field. It is reported that using a classical supercomputer to solve the TFIM problem would take millions of years and require an enormous amount of energy that exceeds global electricity consumption. However, the quantum annealing processor resolved the TFIM problem in a single leap.

The quantum annealing processor does not provide solutions to complex problems all at once; instead, it finds answers through repeated simulations. Quantum annealing is primarily specialized in finding the lowest energy state, that is, the most stable state.

Existing classical methods have limitations when they get trapped in other stable states instead of a solution, making further progress impossible. However, quantum annealing overcomes this issue by utilizing a quantum phenomenon known as 'tunneling effect.' Even if it gets trapped in a state other than the solution, it can tunnel through this state to reach the actual solution. While its application area is more limited than that of quantum computers using logical gate models, its commercialization could be faster.

Bang Jeong-ho, director of the Quantum Computing Center at Yonsei University, noted that 'while traditional quantum computers construct circuits, appropriately place logical gates, implement algorithms, and perform calculations to solve problems, quantum annealing transforms specific problems into 'finding the state with the lowest energy' and solves them,' adding that 'this research demonstrated that it can solve actual problems better than classical computers.'

The TFIM problem solved in this research is closely related to quantum annealing technology. D-Wave's quantum annealing system fundamentally operates based on the principles of TFIM. Therefore, studying TFIM goes beyond analyzing simple physical phenomena; it is also an experimental process to empirically verify how effective quantum annealing actually is. This study is considered an important case that proved this potential.

Andrew King, senior chair scientist at D-Wave, stated that 'successfully solving problems that classical computers find virtually impossible using the quantum annealing processor is an important case that proves the superiority of quantum technology in real scientific applications,' and added, 'there is a great possibility of applying it in solving various optimization problems or complex real-life issues such as artificial intelligence (AI).'

However, Professor Bang Jeong-ho acknowledged that 'while it is true that this research solved a problem significant enough to be published in Science, quantum annealing has not yet proven quantum supremacy for all problems.' He further noted that 'additional research is necessary to confirm whether practical application cases can be resolved in various fields.'

◇D-Wave's claims of quantum supremacy face rebuttal

Some researchers have raised doubts about the research results presented by D-Wave. They argue that since the TFIM problem can also be solved with classical computers, it is difficult to claim quantum supremacy.

Dries Steinhoff, a professor at New York University, reported that his research team solved the TFIM problem in just two hours using a mathematical technique called 'tensor networks' on a standard laptop. Tensor networks are technologies that drastically reduce the computational workload required for simulations.

Andrew King, senior chair scientist at D-Wave, explained that 'this does not imply a complete rebuttal of D-Wave's entire research,' adding that 'they have not solved all the problems we dealt with, nor have they solved problems of the size we studied.' He mentioned that they executed large-scale calculations using up to 3,200 qubits, but the research results have not yet been disclosed.

In response, Professor Steinhoff stated, 'Our approach can be linearly expanded to larger problem sizes, so there is no need to solve bigger problems.' He dismissed the idea that D-Wave could attempt this, arguing that it would be meaningless.

Researchers at École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland also commented that the problems solved by D-Wave do not necessarily require quantum entanglement and can be simulated using classical computers. They claimed to have solved a problem that D-Wave asserted would take 200 years with classical supercomputers in just three days using only four graphics processing units (GPUs).

Giuseppe Carleo, an EPFL professor leading the research, suggested that 'it should be possible to exceed the size of the problem solved by D-Wave within a week.' In response, D-Wave stated, 'In our paper, we already noted that this type of simulation is too simple to assign much significance,' adding that 'while it is true that the EPFL research results represent technological advancement, it does not overturn our claims.'

This controversy extends the ongoing debate over the supremacy of quantum computing. Google claimed in 2019 that its Sycamore quantum computer solved operations that would take classical supercomputers 10,000 years in mere minutes; however, in 2022, it resolved the same operation in 15 hours using 512 GPUs. Last year, a study emerged that solved the same problem in 14.22 seconds. During this process, the tensor network technique played a significant role.

Alex Kissinger, a professor at the University of Oxford, remarked that 'D-Wave is a pioneering corporation that has been developing commercial quantum computers since 2011, but there were questions in the early stages about whether it was a true quantum computer.' He noted that 'while such controversies have diminished, doubts remain as to whether quantum computers can solve problems that classical computers cannot.'

References

Science (2025), DOI: https://doi.org/10.1126/science.ado6285