Supplementary Information for:
Efficient Mean-Field Simulation of Quantum
Circuits Inspired by Density Functional Theory
Marco Bernardi
∗
,
†
†
Department of Applied Physics and Materials Science, California Institute of Technology,
Pasadena, CA 91125, USA.
‡
Department of Physics, California Institute of Technology, Pasadena, CA 91125, USA.
E-mail:
bmarco@caltech.edu
1
1
T
S
H
S
H
H
S
S
H
H
T
S
H
H
T
T
S
S
T
S
H
H
S
H
S
H
H
T
T
S
T
H
T
S
S
S
S
S
H
T
H
S
H
S
S
T
T
T
Figure S1 | Additional QC-DFT simulations of Clifford+T random QCs using
LPA gate functionals.
Random QC using a universal Clifford+T gate set. This type of
QC alternates one step where single-qubit gates chosen at random within the set are applied
to all qubits, and one step where CNOT gates with randomly chosen control-target pairs
are applied to all qubits. This two-step sequence is shaded in gray (top). The accuracy of
the simulated SQPs at each step is shown for this type of random QCs, with a depth of 20
steps, for different numbers of qubits (bottom). Each SQP accuracy curve is obtained by
averaging results from 20 distinct random QC instances.
2
Figure S2 | Additional results for optimized MGA gate functionals.
Accuracy
comparison for the LPA and MGA-3 functionals applied to the Clifford+T QCs in Fig.
S1
.
The main improvements for the MGA-3 over the LPA functional occur between steps 1 and
7, where the multi-gate corrections are applied. These results, shown for QCs with a size of
20 qubits, are obtained by averaging over the same number of QCs as in Fig.
S1
.
3