Limits on the speed of quantum algorithms are not quite as rigid as previously thought, according to new research
Quantum computers promise to simulate physical systems far beyond the reach of classical machines. But even quantum algorithms face fundamental limits: no fast-forwarding theorems. These say that if you want to simulate a system for longer, the computational effort increases at least in proportion.
In many real algorithms, the situation is even worse than the ideal case. This is because the requirements of accuracy and runtime compound one another, so that running longer and running precisely become entangled and costly together.
In their new study, a team of researchers from Beijing and Hong Kong showed that this link between time and cost is not as rigid as it seems, at least for a broad class of realistic systems that interact with their surroundings. These so‑called open quantum systems are described by Lindbladian dynamics, which include both ordinary evolution and the effects of noise or dissipation.
In earlier approaches, improving the accuracy of a simulation made long-time calculations much more demanding, because the two effects multiplied together. The new method separates these contributions, so that the cost of simulating longer times does not automatically amplify the cost of achieving accuracy. In practical terms, this means that extending a simulation in time becomes far more manageable than before.
An even more surprising result appears when the dissipative processes in the system have a particular structure common in quantum physics. In these cases, the number of sequential computational steps needed grows only very slowly as the simulated time increases. Instead of needing proportionally more effort to simulate longer dynamics, the algorithm can effectively skip ahead, compressing what would normally be a long sequence of steps into a much shorter one. This fast‑forwarding is a dramatic departure from the usual expectation that longer simulations must always take longer to run.
The team also demonstrated how these ideas can be applied to studying thermal properties of quantum systems. These are essential for understanding matter at finite temperature. Such calculations are especially challenging at low temperatures, where lots of interesting physics often occurs. By exploiting their more favourable scaling, the new approach can extract certain thermal properties more efficiently, making it easier to probe regimes that are typically hard to access.
Beyond the specific application of this advance itself, the work highlights an important shift in perspective. Rather than focusing only on the worst‑case limits imposed by general theorems, it shows that carefully chosen physical structure can loosen those limits. Only time will tell what the implications for quantum simulation and computation will be in the future.