Methods for in-situ characterization of gaussian boson sampling (GBS) devices
A method includes causing activation, at a first time, of a first set of squeezed light sources from a plurality of squeezed light sources of a Gaussian boson sampling (GBS) circuit. At a second time after the first time, a first photon statistic is detected at a first output port from a plurality of output ports of the GBS circuit. At a third time after the first time, a second set of squeezed light sources from the plurality of squeezed light sources of the GBS circuit is activated, the second set of squeezed light sources being different from the first set of squeezed light sources. At a fourth time after the third time, a second photon statistic is detected at a second output port from the plurality of output ports of the GBS circuit. At least one transformation matrix is estimated that represents a linear optical interferometer of the GBS circuit based on the first photon statistic and the second photon statistic.
1 . A method, comprising:
causing activation, at a first time, of a first set of squeezed light sources from a plurality of squeezed light sources of a Gaussian boson sampling (GBS) circuit;
detecting, at a second time after the first time, a first photon statistic at a first output port from a plurality of output ports of the GBS circuit;
causing activation, at a third time after the first time, of a second set of squeezed light sources from the plurality of squeezed light sources of the GBS circuit, the second set of squeezed light sources being different from the first set of squeezed light sources;
detecting, at a fourth time after the third time, a second photon statistic at a second output port from the plurality of output ports of the GBS circuit; and
estimating at least one amplitude of a transformation matrix that represents a linear optical interferometer of the GBS circuit based at least in part on the first photon statistic and the second photon statistic.
2 . The method of claim 1 , wherein at least one of the first output port of the GBS circuit or the second output port of the GBS circuit is optically coupled to a photon number resolving (PNR) detector.
3 . The method of claim 1 , wherein at least one of the first photon statistic or the second photon statistic includes a photon count and does not include phase information.
4 . The method of claim 1 , further comprising:
causing sequential activation of each squeezed light source from the plurality of squeezed light sources, during a first time period preceding the first time;
measuring, during the first time period and after each sequential activation, a mean photon number at the plurality of output ports, the mean photon number after each sequential activation collectively defining a plurality of mean photon numbers; and
determining, during the first time period, a set of amplitudes of the at least one transformation matrix based on the plurality of mean photon numbers.
5 . The method of claim 4 , wherein determining the set of amplitudes of the at least one transformation matrix is further based on squeezing parameters of the plurality of squeezed light sources.
6 . The method of claim 1 , wherein each squeezed light source from the plurality of squeezed light sources has a fixed configuration.
7 . The method of claim 1 , wherein the plurality of squeezed light sources includes degenerate squeezed light sources.
8 . The method of claim 1 , wherein the plurality of squeezed light sources includes twin-beam squeezed light sources.
9 . The method of claim 1 , wherein at least one of the first set of squeezed light sources or the second set of squeezed light sources is selected based on an expected amplitude of the transformation matrix.
10 . The method of claim 1 , wherein at least one squeezed light source from the plurality of squeezed light sources is configured to generate light having a single-mode state.
11 . The method of claim 1 , wherein at least one squeezed light source from the plurality of squeezed light sources is configured to generate light having a two-mode state.
12 . The method of claim 1 , wherein the estimating the at least one amplitude is further based on an additional photon statistic.
13 . The method of claim 1 , further comprising:
activating, subsequent to the fourth time, at least one additional set of squeezed light sources of the GBS circuit, the at least one additional set of squeezed light sources different from the first set of squeezed light sources and the second set of squeezed light sources; and
detecting at least one additional photon statistic, via at least one additional output port from the plurality of output ports of the GBS circuit.
14 . The method of claim 1 , further comprising:
determining, subsequent to the fourth time, that an element of the transformation matrix remains to be estimated; and
in response to the determining that the element of the transformation matrix remains to be estimated:
activating at least one additional set of squeezed light sources of the GBS circuit, and
detecting at least one additional photon statistic, via at least one additional output port from the plurality of output ports of the GBS circuit.
15 . The method of claim 1 , wherein each squeezed light source from the plurality of squeezed light sources has a fixed configuration, and the plurality of squeezed light sources includes at least one of a degenerate squeezed light source or a twin-beam squeezed light source.
16 . The method of claim 1 , wherein each squeezed light source from the plurality of squeezed light sources has a fixed configuration associated with a squeezing value of that squeezed light source.
17 . The method of claim 1 , wherein each output port from the plurality of output ports of the GBS circuit is optically coupled to a photon number resolving (PNR) detector.
18 . The method of claim 1 , wherein the transformation matrix includes a substantially unitary matrix.
19 . The method of claim 1 , further comprising estimating a plurality of phases of the transformation matrix based on at least one correlation function.
20 . The method of claim 1 , wherein the estimating the at least one amplitude of a transformation matrix is based on a correlation function.