Quantum Annealing
A quantum optimization approach that seeks low-energy states by slowly evolving a quantum system.
Definition
Quantum annealing solves optimization problems by encoding them as the energy landscape of a quantum system, then slowly evolving the system so it settles into a low-energy state that represents a good solution. Quantum tunneling can help escape local minima that trap classical methods.
Annealers are far more numerous in qubit count than gate-model machines but far more limited in what they compute, and independent studies have struggled to show a clear, general speedup over the best classical heuristics. Their value remains problem-specific and actively debated.
The commercial availability of large annealers has produced a substantial body of empirical study, and the consistent finding is that a clear, general advantage over the best classical heuristics has been hard to establish. Advantages, where reported, tend to be problem-specific and sensitive to how the problem is mapped onto the hardware's limited connectivity. This makes annealing a case study in the care needed to evaluate quantum performance claims.
Contrast with gate model
- Specialized for optimization, not universal computation.
- Uses continuous evolution rather than discrete gates.
- Realized commercially at large qubit counts, though with limited connectivity.
Why it matters
Annealing targets the combinatorial optimization problems common in scheduling, logistics, and design. Whether it offers a real advantage over strong classical solvers such as simulated annealing remains an open, problem-dependent question.
Fusion connection
Combinatorial layout and scheduling problems in engineering are candidate annealing applications, evaluated by Kronos against classical optimizers on the same problem before any adoption.