New way to simulate quantum sampling with noise
Researchers developed a framework to classically simulate boson and fermion sampling when thermal loss is present.
Scientists developed a framework to classically simulate boson and fermion sampling under thermal loss. This includes cases of pure loss.
The work covers Fock-state boson sampling, Gaussian boson sampling with fixed uniform squeezing, and fermion sampling with magic inputs. These experiments are a leading approach to demonstrating quantum computational advantage.
Sampling experiments are used to show quantum computational advantage. Thermal loss refers to noise in the system. Haar-random interferometers are used in these experiments.
For certain types of boson and fermion sampling, the simulation runs in polynomial-time. This occurs with an inverse-polynomial total variation error for any fixed nonzero overall loss rate. This happens with high probability over Haar-random interferometers.
A milder flatness assumption on the ideal output distribution allows for quasipolynomial-time simulation without Haar randomness. With constant thermal loss per layer, this same flatness assumption allows polynomial-time simulation at logarithmic depth.
The bounds for Fock-state boson sampling cover both dilute and saturated regimes. When only a bounded number of input modes are empty, the noisy output converges to a distribution independent of the interferometer.
Key facts
- The framework allows for polynomial-time classical simulation of boson and fermion sampling under thermal loss.
- Simulation is possible at inverse-polynomial total variation error for any fixed nonzero overall loss rate.
- The bounds for Fock-state boson sampling cover both dilute and saturated regimes.