Fig. 2: Circuit diagram for kernel function.
From: Machine learning of high dimensional data on a noisy quantum processor
a 14-qubit example of the circuit used for experiments in this work. The dashed line indicates the boundary between U(xi) and U†(xj), which are run sequentially to sample ∣〈ϕ(xj)∣ϕ(xi)〉∣2. Non-virtual gates occurring at the boundary are contracted for hardware runs. b The basic encoding block consists of a Hadamard followed by three single-qubit rotations, each parameterized by a different element of the input data x (normalization and encoding constants omitted here). c We used the \(\sqrt{\,{{\mbox{iSWAP}}}\,}\) entangling gate, a hardware-native two-qubit gate on the Sycamore processor.
