www.nature.

com/scientificreports

OPEN Non‑adaptive measurement‑based

quantum computation on IBM Q
Jelena Mackeprang 1,2,3, Daniel Bhatti 1,2 & Stefanie Barz 1,2*

We test the quantumness of IBM’s quantum computer IBM Quantum System One in Ehningen,
Germany. We generate generalised n-qubit GHZ states and measure Bell inequalities to investigate
the n-party entanglement of the GHZ states. The implemented Bell inequalities are derived from
non-adaptive measurement-based quantum computation (NMQC), a type of quantum computing
that links the successful computation of a non-linear function to the violation of a multipartite Bell-
inequality. The goal is to compute a multivariate Boolean function that clearly differentiates non-local
correlations from local hidden variables (LHVs). Since it has been shown that LHVs can only compute
linear functions, whereas quantum correlations are capable of outputting every possible Boolean
function it thus serves as an indicator of multipartite entanglement. Here, we compute various non-
linear functions with NMQC on IBM’s quantum computer IBM Quantum System One and thereby
demonstrate that the presented method can be used to characterize quantum devices. We find a
violation for a maximum of seven qubits and compare our results to an existing implementation of
NMQC using photons.

Commercially available quantum computers (QCs) have arrived in the NISQ (noisy intermediate-scale quantum)
­era1. Equipped with 10s to 100s of of noisy qubits, these devices already enable the implementation of quantum
operations and thus basic quantum ­algorithms2. Despite the lack of error correction, algorithms and techniques
adapted to the strengths and shortcomings of the computers could facilitate non-classical computation within
the near future. To compare the performance of the large range of different quantum devices and to find the
best-suited QC for a specific problem, benchmarking, i.e., reproducibly measuring the performance of quantum
devices, becomes especially ­important3.
To be independent of the architecture and capture the complexity of quantum machines, benchmarking
protocols go beyond comparing the various hardware c­ haracteristics4,5. The goal is to find protocols that give
maximal information about the performance of a quantum ­device2. Examples for such hardware benchmarks
are randomised b ­ enchmarking6, cross-entropy b ­ enchmarks7, or the quantum v­ olume4,8. Besides that, application
benchmarks exist which test the performance of NISQ devices based on their execution of different applications
or algorithms and help one to understand how good QCs can deal with different ­tasks2,5.
One fundamental type of application that can be used to benchmark quantum devices is the generation and
verification of e­ ntanglement9–18. To this aim, various tests of multipartite entanglement have been implemented,
e.g. utilising Mermin i­nequalities9–12 or multiparty Bell ­inequalities13,14, but also measuring the entanglement
between all connected qubits in a large graph ­state15,16. In the case of Greenberger-Horne-Zeilinger (GHZ) states
a feasible method to estimate the fidelity has been derived and implemented to verify the state generation of
large numbers of q ­ ubits17,18.
In this work, we make use of a method called non-adaptive measurement-based quantum computation
(NMQC) to characterise an IBM QC with 27 superconducting qubits.
The goal in NMQC is to compute a multivariate function. While local hidden variables (LHVs) can only
output linear functions, quantum correlations can compute all Boolean functions. The success of such a com-
putation can be related to the violation of a (generalised) Bell inequality and proves the advantage over classical
­resources19. So far, binary NMQC has been implemented with four-photon GHZ s­ tates20. Here, we use GHZ states
on an IBM QC to implement NMQC with more than four qubits. This allows us to test the quantum correlations
of the generated GHZ states and therefore the non-classicality of the respective IBM QC.
In particular, we implement NMQC for one two-bit function, three three-bit functions, and one four-bit, one
five-bit, and one six-bit function on the superconducting quantum computing system IBM Quantum System
One (QSO) and demonstrate that it exhibits multipartite entanglement. For qubit numbers lower or equal to

1
Institute for Functional Matter and Quantum Technologies, University of Stuttgart, 70569 Stuttgart,
Germany. 2Center for Integrated Quantum Science and Technology (IQST), University of Stuttgart,
70569 Stuttgart, Germany. 3Present address: QuSoft and Centrum Wiskunde & Informatica (CWI), Science Park
123, 1098 XG Amsterdam, The Netherlands. *email: [email protected]

Scientific Reports | (2023) 13:15428 | https://doi.org/10.1038/s41598-023-41025-4 1

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Vol.:(0123456789)
 www.nature.com/scientificreports/

five, we utilise quantum readout error ­mitigation21 to reduce noise from local measurement errors. For higher
qubit numbers, we utilise the error mitigation tools provided by Q ­ iskit22. We demonstrate violations of the
associated Bell inequalities for up to seven qubits, which indicates the non-classical properties of the quantum
computing system.

Background
NMQC. First, let us briefly describe the general idea of NMQC (for a detailed overview over the proce-
dure see Fig. 1 and, e.g. Refs.19,23): starting from a classical n-bit input string x = (x0 , x1 , . . . , xn−1 ) ∈ {0, 1}n,
which is sampled from a probability distribution ξ(x), the goal is to compute any multivariate Boolean function
f : {0, 1}n → {0, 1}. For this, one has access to a restricted classical computer limited to addition mod 2, which
can be used for classical pre- and post-processing (see Fig. 1). The core of NMQC is embodied by non-adaptive
measurements on an l-qubit resource state, with l ≥ n. It has been shown that if the measurement statistics are
described by local hidden variables (LHVs)24, i.e. one uses a classical resource state, the output of NMQC is
restricted to linear functions. As the pre-processor is already capable of outputting linear functions LHVs thus
do not “boost” the pre-processor’s computational ­power19.
In contrast, non-local quantum correlations, can elevate the pre-processor to classical universality. The gen-
eralised l-qubit GHZ state
1  
|GHZ� = √ |0�⊗l + |1�⊗l , (1)
2
enables the computation of all functions f : {0, 1}n → {0, 1} with at most l = 2n − 1 qubits. The computation of
a non-linear function requires non-locality25 and can be seen as a type of GHZ ­paradox19. This means that the
successful execution of NMQC demonstrates non-locality. Note that in our case the non-locality is realised by
single-qubit measurements on a multipartite entangled state and we thus use the two terms non-locality and
multipartite entanglement interchangeably.
In general, it can be shown that the average success probability p̄S = p(z = f (x)), i.e. the probability that
the output z is identical to the value of the target function f(x), is related to a normalised Bell inequality β with
a classical (LHV) bound βc and a quantum bound βq19:

 βc
2p̄S − 1 = β = (−1)f (x) ξ(x)E(x) ≤ . (2)
βq
x

The expectation values are defined:

E(x) = p(z = 0|x) − p(z = 1|x), (3)
where p(z = k|x) is the probability that z is equal to k for the input x.
It has been shown that the GHZ state always maximally violates the given Bell inequalities and minimises
the number of required qubits for a ­violation26,27. It is thus optimal for ­NMQC19 and we will use it as a resource
in the following investigations on IBM QSO.

Tested functions and Bell inequalities. The NMQC computations for the four-qubit GHZ state pre-
sented in this work result from the two-variate function:

Input: ( bits) Output:

meas. meas.
settings: results:
( bits) ( bits)

Correlated resource ( -qubit GHZ state)

Figure 1.  The figure shows the general scheme of NMQC. At the beginning, an input string x ∈ {0, 1}n is sent
to the parity computer, which in turn computes the bit string s ∈ {0, 1}l . This restricted computation can be seen
as a matrix vector multiplication: s = (Px)⊕, where P is an l-by-n binary matrix and ⊕ denotes that the matrix
vector product is evaluated w.r.t. mod 2 operations. Each bit si ∈ {0, 1} in s now determines the settings for the
measurement on the ith qubit of the l-qubit resource state. For each subsystem, there are two measurement
operators m̂i (si ), one can choose from (here m̂i (0) ≡ X and m̂i (1) ≡ Y , with X and Y denoting the Pauli
operators). Each measurement yields one of two possible measurement results Mi ∈ {−1, 1}, which can be
mapped to bits mi ∈ {0, 1}: Mi = (−1)mi. The measurements are performed on a  correlated l-qubit resource and
the measurement results mi ∈ {0, 1} are summed up by the parity computer: z ≡ i mi. Finally, if z = f (x) for
this input x, the computation was successful. Note that if z = f (x) for every x, we say that an NMQC scheme is
deterministic. The figure has been adapted from Ref.23.

Scientific Reports | (2023) 13:15428 | https://doi.org/10.1038/s41598-023-41025-4 2

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Vol:.(1234567890)
www.nature.com/scientificreports/

NAND2 (x) = x0 x1 ⊕ 1, (4)

and the three three-variate functions:
h3 (x) = x0 x1 ⊕ x0 x2 ⊕ x1 x2 ⊕ x0 ⊕ x1 ⊕ x2 , (5)

OR3 (x) = x0 ∨ x1 ∨ x2 , (6)

OR⊕
3 (x) = OR3 (x) ⊕ x0 x2 , (7)
where ∨ is the logical OR operator and ⊕ denotes addition mod 2. Note that the same functions have been used
to implement NMQC using a four-photon GHZ state in Ref.20.
In the case of the two-bit function NAND2 (x), we use a uniform probability distribution ξ(x) = 41 , which
yields the Bell inequality [see Eq. (2)]:

1 βc = 1/2
βNAND2 = [−E((0, 0)) − E((1, 0)) − E((0, 1)) + E((1, 1))] ≤
βq = 1
. (8)
4
The relation between the measurement settings and the measurements, i.e., m̂i (si ) = X/Y for si = 0/1
(i ∈ {0, 1, 2, 3}) allows us to rewrite the Bell inequality (8) in terms of the four measurements:

1 βc = 1/2
βNAND2 = �−XXYY − YXXY − XYXY + YYYY � ≤
βq = 1
, (9)
4
where we additionally made use of the following relation between measurement settings and the input bits xi:
s0 = x0 , s1 = x1 , s2 = x0 ⊕ x1 ⊕ 1, s3 = 1. (10)
In the same manner one finds the Bell inequalities for the three three-variate functions given in Eqs. (5–7)20. The
inequalities and the respective pre-processing implemented in this paper are shown in Table 1.
To perform NMQC for five- to seven-qubit GHZ states we use the generalisation of h3 (x), namely hk (x), for
k = 4, 5 and 6:
 
k−2
� k−1
� k−1
�
hk (x) = xi  xj  ⊕ xi . (11)
i=0 j=i+1 i=0

For any k the sampling distribution is uniform, i.e. ξ(x) = 1/2k , and the pre-processing is given by:
xi i = 0, . . . , k − 1

si = k−1 . (12)
j=0 j i = k
x

The Bell inequality induced by hk (x) and defined by the pre-processing (12) and the uniform sampling distri-
bution has the quantum bound βq = 1. This can be seen by explicitly computing all expectation values. The
classical bounds can either be found numerically or inferred from the connection between
−k
NMQC and classical
 
Reed–Muller error-correcting codes, as pointed out in Ref.28. They are equal to βc = 2 2 for even k and 2− 2
k−1

for odd k. We elaborate on this in the Supplementary Information and further show that in order to compute
the k-bit function hk (x) with NMQC, one only requires k + 1 qubits.

NMQC on IBM Quantum System One. IBM QSO in Ehningen, Germany, is a 27-qubit QC, which we
used to run NMQC for up to seven qubits. Testing a possible violation of Bell inequalities for different qubit
configurations of the QC allows for a characterisation of the whole QC or a subset of qubits.

x (0, 0, 0) (1, 0, 0) (0, 1, 0) (0, 0, 1) (1, 1, 0) (1, 0, 1) (0, 1, 1) (1, 1, 1)

�m̂0 (s0 )m̂1 (s1 )m̂2 (s2 )m̂3 (s3 )� XXXX YXXY  XYXY  XXYY  YYXX YXYX XYYX YYYY  βc βq
(−1)OR3 (x) ξ(x) +3/10 −1/10 −1/10 −1/10 −1/10 −1/10 −1/10 −1/10 4/10 8/10
9/16 14/16
⊕
(−1)OR3 (x) ξ(x) +1/16 −3/16 −3/16 −1/16 −3/16 +1/16 −3/16 +1/16

(−1)h3 (x) ξ(x) +1/8 −1/8 −1/8 −1/8 −1/8 −1/8 −1/8 +1/8 1/2 1

Table 1.  Bell inequalities for the three three-variate functions OR3 (x), OR⊕ 3 (x) and h3 (x) given in Eqs. (5–7)
and implemented in the “Results” section. The first row shows all possible three-bit inputs, while the second
row gives the respective measurements after pre-processing. The pre-processing used for all three functions
is: s0 = x0, s1 = x1 , s2 = x2 , s3 = x0 ⊕ x1 ⊕ x2. m̂i (si ) = X/Y for si = 0/1 (i ∈ {0, 1, 2, 3}). Rows 3–5 show the
resulting prefactors of the different measurement results in the Bell inequalities. The classical (quantum) bound
βc (βq) of each Bell inequality are presented in the two separate columns on the right.

Scientific Reports | (2023) 13:15428 | https://doi.org/10.1038/s41598-023-41025-4 3

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Vol.:(0123456789)
 www.nature.com/scientificreports/

The QC’s architecture is shown in Fig. 2, where each qubit (vertex) is marked by its physical qubit number and
edges denote physical connections between qubits. Here, physical connection means that two qubits are directly
coupled, which allows for a direct implementation of two-qubit gates between those qubits. In the following,
when mentioning the physical qubit numbers, we refer to the numbering depicted in Fig. 2. At the time of the
experiment, the QC contained a Falcon r5.11 processor and its backend version was 3.1.9.
We perform two experiments on IBM QSO: (i) the physical qubits are chosen by Qiskit and the quantum cir-
cuit is optimised by Qiskit, and (ii) the physical qubits are chosen manually and the quantum circuit is optimised
by our own method (see “Methods” section for details). In both experiments the goal is to generate generalised
GHZ states [see Eq. (1)] as a resource to perform NMQC.
While in (i) we only test a single configuration, i.e. the one chosen by Qiskit, in (ii) we generate and test every
possible l-qubit configuration, where l is the number of qubits. By “qubit configuration”, we mean the collection
of l physical qubits that are physically connected in the QC (see Fig. 2). For each tested Bell inequality in (ii) we
then average over all measured bounds for the distinct distributions to determine the measured bound of the
whole QC.

Error mitigation. We post-process the measured data for up to five qubits, using the quantum readout error
mitigation (QREM)21. This method has been used, for example, in Ref.18, where it has led to considerable
improvements in the fidelity of a generated multi-qubit GHZ state. It aims at mitigating readout errors, which
are errors during the measurement of the state of a single qubit and the main assumption is that these measure-
ment errors are local. We explain the details in the “Methods” section.
To improve the results for NMQC using six and seven qubits, QREM seems to be insufficient. In fact, we
observed a negative effect on the measured bounds and thus switch to the measurement error mitigation (MEM)
provided by ­Qiskit22. In contrast to QREM Qiskit’s MEM does not assume measurement errors to be local but
global. This means that instead of n 2 × 2 calibration matrices Ai one needs to determine a single 2n × 2n calibra-
tion matrix A by preparing and measuring all 2n basis states (see Ref.22 for details).

Results
Here, we present the average values for the violations of the associated Bell inequalities of NMQC listed above. We
start with the first experiment (i), in which the physical qubits are chosen by Qiskit. There, the circuits for NMQC
were all transpiled using the option “optimisation level 3”, i.e. heavy optimisation including noise-adaptive qubit
mapping and gate ­cancellation22. We differentiate between two sub-experiments: one, where the circuits were
transpiled using the option “layout_method=dense”, which chooses the most connected subset of qubits with
the lowest noise and one, where the circuits were transpiled using the option “layout_method=noise_adaptive”,
which tries to map the virtual to physical qubits in a manner that reduces the ­noise22.
In the second experiment (ii) we choose the qubits manually, testing all possible qubit configurations to
generate the l-qubit GHZ states and perform NMQC.

Experiment (i): transpilation optimisation level 3. Figure 3 shows the measured bounds of the Bell
inequalities for optimisation level 3 and two different layout methods. Each Bell inequality was tested in 70 sepa-
rate runs, where in each run every circuit, induced by the respective function, has been executed 1000 times, i.e.
1000 runs. One can see that for both methods all measured values, except for h6 (x), i.e. seven qubits, are above
the classical bounds which translates to a quantum advantage in the associated NMQC games, even when taking
into consideration the standard deviations determined from the 70 runs. This, in turn, means that the quantum
average success probability of the probabilistic NMQC games is higher than the LHV one, indicating multipar-
tite entanglement. For this experiment one can say that the performance of both methods “dense” and “noise
adaptive” provided by Qiskit was similar.

Experiment (ii): transpilation optimisation level 0 and error mitigation. Figure 4a shows the
measured bounds of the Bell inequalities averaged over all possible qubit configurations for four (OR3 (x),
3 (x), NAND2 (x) and h3 (x)), five (h4 (x)), six (h5 (x)), and seven (h6 (x)) qubits and the mitigated bounds
OR⊕

(a) 6 17 (b) 6 17

Q0 Q1 Q4 7 10 12 15 18 21 23 0 1 4 7 10 12 15 18 21 23

Q2 13 24 2 13 24

3 5 8 11 14 16 19 22 25 26 3 5 Q8 Q11 Q14 Q16 Q19 22 25 26

9 20 Q9 20

Figure 2.  Architecture of IBM Quantum System One for different examples of l-qubit configurations. The
nodes indicate qubits, marked by the physical qubit numbers and the edges denote which ones are physically
connected. The qubit configurations are marked in red and the respective qubits are labelled by Qk, where k is
the physical qubit number. (a) 4-qubit configuration 0–1–2–4. (b) 6-qubit configuration 8–9–11–14–16–19.

Scientific Reports | (2023) 13:15428 | https://doi.org/10.1038/s41598-023-41025-4 4

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Vol:.(1234567890)
www.nature.com/scientificreports/

(a) 1.1 (b) 1.1

βq βmeas βc βq βmeas βc
0.9 0.9

0.7 0.7

0.5 0.5
β

β
0.3 0.3

0.1 0.1
0.0 0.0
−0.1 −0.1
OR3 OR⊕
3
NAND2 h3 h4 h5 h6 OR3 OR⊕
3
NAND2 h3 h4 h5 h6

Figure 3.  Measured bounds of the Bell inequalities, averaged over 70 runs with 1000 shots per circuit, induced
by the functions OR3 (x), OR⊕ 3 (x), NAND2 (x) and hk (x) for 3 ≤ k ≤ 6 and their standard deviations for (a)
optimisation level 3 and the “dense” layout method and (b) optimisation level 3 and the “noise adaptive”
layout method. The red (diagonally striped \\)/grey (dotted) bars denote the theoretically achievable quantum/
classical bounds and the white (diagonally striped //) bars stand for the measured values. The exact values for all
measured bounds are listed in Table 2.

(a) 1.0 (b)

βq corr
βmeas βmeas βc 1.0 q max
corr
max c

0.8 0.8

0.6 0.6
β

0.4 0.4

0.2 0.2

0.0 0.0
OR3 OR⊕
3
NAND2 h3 h4 h5 h6 OR3 OR3 NAND2 h3 h4 h5 h6

Figure 4.  (a) Measured average bounds of the Bell inequalities induced by the functions OR3 (x), OR⊕ 3 (x),
NAND2 (x) and hk (x) for 3 ≤ k ≤ 6 and their standard deviations for optimisation level 0 and the mitigated
bounds. The red (diagonally striped \\)/grey (dotted) bars denote the theoretically achievable quantum/classical
bounds and the white (diagonally striped //) bars stand for the measured values. The orange (plain) bars denote
mitigated bounds. (b) Measured and mitigated bounds of the qubit configuration which produced the highest
values, induced by the same functions for optimisation level 0. The exact values for all measured bounds as well
as the qubit configurations are listed in Table 2.

improved by error correction. The error correction techniques applied are QREM (four and five qubits) and
Qiskit’s integrated MEM (six and seven qubits). For every qubit configuration there is exactly one NMQC run
with 1000 shots per circuit. Note that, if the calibration data of the backend had changed during the NMQC run,
the data was discarded and the run repeated. Note further that before every run, the data needed for the error
mitigation was generated. In Fig. 4b we show the measured bounds of the qubit configuration, which produced
the highest violation (exact values and the physical qubit numbers are shown in Table 2).
One can see from the plots and the data (see Fig. 4a, b and Table 2) that not only the configurations, which
produced the highest values, but also the averaged results are significantly higher than the classical bounds of the
respective Bell inequalities for any tested function and number of qubits. Especially for more than three qubits
the averaged values are higher than in the case of letting Qiskit choose the qubit configuration (see Fig. 3 and
Table 2). In the case of the single configurations one should keep in mind that these results only express a single
run (see Supplementary Information for details).
However, it is important to note that the performance of the QC strongly varies depending on the configura-
tion and the time of execution (see Supplementary Information). This explains why performance differed across
the four four-qubit experiments. For example, OR3 (x) performed significantly better than OR⊕ 3 (x).
The presented error margins correspond to the 99% confidence intervals of the measured values with respect
to 1000 bootstrapped samples for each function except for h6 (x), where we used 100 bootstrapped samples. We
chose ­bootstrapping29 instead of sampling at different moments in time as the performance of the quantum
processor varied considerably. Due to the heavy bias introduced by the optimization procedure used in error
mitigation, we omitted error margins as statistical errors are not a meaningful measure in this situation.

Scientific Reports | (2023) 13:15428 | https://doi.org/10.1038/s41598-023-41025-4 5

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Vol.:(0123456789)
 www.nature.com/scientificreports/

OR3 (x) OR⊕

3 (x) NAND2 (x) h3 (x) h4 (x) h5 (x) h6 (x) Opt. level Layout Figure
βmeas 0.593 ± 0.042 0.670 ± 0.030 0.757 ± 0.040 0.846 ± 0.010 0.426 ± 0.036 0.339 ± 0.040 0.056 ± 0.030 3 Dense 3a
βmeas 0.614 ± 0.034 0.767 ± 0.017 0.837 ± 0.013 0.760 ± 0.026 0.683 ± 0.026 0.271 ± 0.061 0.032 ± 0.124 3 Noise 3b
βmeas 0.686 ± 0.002 0.618 ± 0.003 0.725 ± 0.004 0.834 ± 0.003 0.629 ± 0.002 0.484 ± 0.001 0.306 ± 0.001 0 All 4a
corr
βmeas 0.753 0.700 0.806 0.911 0.715 0.566 0.366 0 All 4a
βmax 0.743 ± 0.012 0.796 ± 0.013 0.907 ± 0.017 0.906 ± 0.012 0.845 ± 0.011 0.837 ± 0.008 0.774 ± 0.006 0 Best 4b
corr
βmax 0.792 0.869 0.992 0.979 0.932 0.935 0.871 0 Best 4b

Table 2.  Measured averaged values and the standard deviations for the bounds of the Bell inequalities induced
by the NMQC games listed in the “Background” section. Results are shown for different Qiskit optimisation
levels, Qiskit layout methods and manually chosen qubit mapping, without (βmeas) and with (βmeas
corr ) error
mitigation. For the layout method “best” the exact qubit configurations are: 1–2–4–7 (OR3 (x)), 0–1–2–3
(OR⊕ 3 (x)), 16–19–14–22 (NAND2 (x)), 24–18–21–23 (h3 (x)), 10–12–13–14–16 (h4 (x)), 4–7–10–12–13–15
(h5 (x)), 11–12–13–14–16–19–20 (h6 (x)).

Comparison to photonic NMQC. In Ref.20 binary NMQC has been implemented using four-photon
GHZ states, testing the functions OR3 (x), OR⊕ 3 (x), NAND2 (x) and h3 (x). Here, we compare our results using
IBM QSO to the photonic results.
In Fig. 5a we show the measured bounds from the photonic experiments and the respective standard devi-
ations. For a better comparison, we calculate the difference between these values and the results of experi-
ment (ii), with and without error mitigation, i.e. �βmeas/max = βmeas/max (photons) − βmeas/max (QSO) and
corr
�βmeas/max corr
= βmeas/max corr
(photons) − βmeas/max (QSO). In Fig. 5b we plot the difference to the measured bounds
averaged over all qubit configurations (see Fig. 4a) and in Fig. 5c we plot the difference to the measured bounds
of the qubit configuration which produced the highest values (see Fig. 4b).
We find that the photonic values are higher than the uncorrected results using IBM QSO comparing to both
the averaged bounds and the highest bounds (except �βmax for OR3 (x)). Using error mitigation the averaged
values come closer to the photonic results but only exceed them in the case of OR3 (x). Only when applying error
mitigation to the highest values produced by a single qubit configuration the photonic results are exceeded for
every function. Additionally, one has to take into account that the values measured on IBM QSO strongly vary
depending on the configuration and the time of execution (see Supplementary Information for details), which
explains the different behaviour for OR3 (x). The possibility to go to larger numbers of qubits remains a big
advantage of IBM QSO.

(a)1.0 (b) 0.3 (c) 0.3

βq βc ∆βmeas corr
∆βmeas ∆βmax corr
∆βmax
βmeas
0.8
0.2 0.2

0.6
0.1 0.1
β

0.4

0.0 0.0
0.2

0.0 −0.1 −0.1

OR3 OR⊕
3
NAND2 h3 OR3 OR⊕
3
NAND2 h3 OR3 OR⊕
3
NAND2 h3

Figure 5.  (a) Measured average bounds of the Bell inequalities induced by the functions OR3 (x), OR⊕ 3 (x),
NAND2 (x) and h3 (x) and their standard deviations for a photonic implementation of NMQC using four-
photon GHZ ­states20. The red (diagonally striped \\)/grey (dotted) bars denote the theoretically achievable
quantum/classical bounds and the white (diagonally striped //) bars stand for the measured values. (b)
and (c) Difference between the photonic results and the results of experiment (ii), with and without error
mitigation, i.e. �βmeas/max = βmeas/max (photons) − βmeas/max (QSO) [white (diagonally striped //) bars] and
corr
�βmeas/max corr
= βmeas/max corr
(photons) − βmeas/max (QSO) [orange (plain) bars]. (b) Difference to the averaged
measured bounds. Without error mitigation the photonic values are always larger and �βmeas is positive. With
error mitigation the differences become smaller, but only for OR3 the mitigated result exceeds the photonic
value. (c) Difference to the highest values produced by a single configuration. Even without error mitigation the
differences �βmax are small, however only for OR3 the photonic value is smaller and �βmax becomes negative.
With error mitigation the results produced by QSO exceed the photonic values and �βmax corr is always negative.

Scientific Reports | (2023) 13:15428 | https://doi.org/10.1038/s41598-023-41025-4 6

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Vol:.(1234567890)
www.nature.com/scientificreports/

Discussion
On average, we have reached violations of all measured Bell inequalities for all tested functions listed in the “Back-
ground” section on the 27-qubit IBM QSO in Ehningen, Germany. In the cases where Qiskit has chosen exactely
one configuration of physical qubits for every experiment, violations have been measured for all functions for
up to six qubits. This is in contrast to the cases where we have tested all possible l-qubit configurations, configu-
ration by configuration, where l is the number of qubits. There, the highest measured success probabilities of
NMQC computations translate to violations of the tested Bell inequalities (with and without error mitigation) for
up to seven qubits. Note that we observe this not only for the optimally performing qubit configuration but also
for the measured success probabilities averaged over all possible qubit configurations. Since we have investigated
every possible qubit configuration and averaged over all results, we have thereby tested the quantumness of the
device. This means that we have shown a computational advantage in terms of NMQC using the device IBM
QSO and thus demonstrated its non-local behaviour for up to seven qubits. Furthermore, we have compared our
results using four qubits to an existing implementation of NMQC using four-photon GHZ s­ tates20.
An obvious question would be if it is possible to generalise NMQC as an indicator of non-locality to higher
qubit numbers. The ratio between the quantum and the classical success probability of computing the k-bit
pairwise AND function hk (x) increases exponentially with the number k of input bits. It also requires only
k + 1 qubits for its deterministic c­ omputation19. It is thus well suited for generalisation to higher input bit and
therefore qubit numbers. To improve the results and carry out NMQC for hk (x) for k > 7, one could apply more
sophisticated error mitigation/correction t­ echniques18. It would also be interesting to find other functions that
translate to convenient Bell inequalities to test the non-classicality of QCs using this computational test. For
this, one could use the relation between NMQC and Reed-Muller codes hinted at in the “Background” section.
However, the performance of the qubits varies widely over time and this variance should definitely be taken into
account in order to obtain larger GHZ state fidelities and thus better results. It is likely that, in the future, more
sophisticated qubit mapping methods will be developed, such a­ s30, which, in combination with error mitigation
and error correction methods, could facilitate NMQC with large numbers of qubits.
Another possibility to reduce errors and noise in the generation of the GHZ states could be to minimise the
depth of the quantum circuit. In Ref.31 a method has been discussed in which GHZ states of arbitrary size can
be generated with constant circuit depth. Although additional ancilla qubits are needed, here the advantage
gained from the constant circuit depth would presumably surpass problems caused by the increased number of
qubits. From the generation of linear graph states on IBM QCs, which also has a constant depth, it is known that
entangled states of much larger size can be ­generated14,16.
In conclusion, we have implemented NMQC for up to seven qubits using a 27-qubit IBM QC. We have
shown that the calculation of non-linear Boolean functions and the simultaneous violation of multipartite Bell
inequalities can be used to characterise quantum devices. This method can easily be extended to different quan-
tum computing systems with qubits but also to higher-dimensional s­ ystems23.

Methods
Creation of the GHZ state. The scheme used to generate the multi-qubit GHZ states in the first experi-
ment (i) follows an easily scalable ­manner18, consisting of a single Hadamard (H) gate and n − 1 CNOT gates
(see Fig. 6). Qiskit chooses the mapping of the virtual to the physical qubits and optimises the quantum circuit
according to its highest optimisation level.
In the second experiment (ii), where we average over all possible qubit configurations, the qubit onto which
the Hadamard gate acts is the one with the largest numbers of neighbours in the configuration and the one with
the smallest readout-error rate. The readout-error rates are obtained from the backend’s calibration data which
is updated before every NMQC run. If the calibration data changed during an NMQC run, the measured data
was discarded and the run repeated. The CNOT gates are arranged in such a way that as many as possible of
them can be carried out simultaneously, which minimises the circuit ­depth18. Note that CNOT gates are only
applied between physically connected qubits.

q0 0 H
q1 0
q2 0

q
q
Figure 6.  Theoretical scheme for the creation of a GHZ state. First a Hadamard (H) is applied on the virtual
qubit q0. Then n − 1 CNOT gates are carried out between the qubits qi (control) and qi+1 (target), where i runs
from i = 0 to i = n − 1.

Scientific Reports | (2023) 13:15428 | https://doi.org/10.1038/s41598-023-41025-4 7

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Vol.:(0123456789)
 www.nature.com/scientificreports/

Quantum readout error mitigation. In this section we explain the details of the quantum read-
out error mitigation (QREM) introduced in Ref.21. It aims to mitigate readout errors, which are errors dur-
ing the measurement of the state of a single qubit. For example, a qubit might actually be in the state
|1�, but the measurement device asserts that it is in the state |0�. The main assumption in QREM is that
these measurement errors are local. This means that the measurement errors act on the probability vector
p� ≡ (p(0, 0, 0, . . . , 0), p(0, 0, 0, . . . , 1), . . . , p(1, 1, 1, . . . , 1))T , where p(mn−1 , mn−2 , . . . , m1 , m0 ) is the probabil-
ity of obtaining the measurement result mi for the measurement (in the computational basis) of the ith qubit q0
(i ∈ {0, 1, . . . , n − 1}), in the following way:
n  
 pi (0|0) pi (0|1)
�p ′ = �
An−i p, Ai ≡ . (13)
pi (1|0) pi (1|1)
i=1

The Ai are called the calibration matrices and pi (x|y) are the probabilities of measuring the state x given that the
ith qubit was actually prepared in the state y.
In order to mitigate the readout errors, one has to first compute the calibration matrices by preparing the
qubits in the various states and then estimating the probabilities pi (x|y) using the law of large numbers. The
corrected probability vector p is then obtained from the experimental probability vector p� ′ by inverting the
calibration matrices. However, sometimes, as the estimation of Ai is not exact, the resulting p may not be an
actual physical probability vector, meaning that some element of it may be smaller than 0 or all of them do not
sum up to 1. Therefore, if that is the case, we use an optimisation method to find the closest physical probability
vector p� ∗ to p . To be exact, p� ∗ is given ­by21:

p� ∗ = arg min
(�p�˜ − p��), (14)
∀i p̃i ≥0, i p̃i =1

where � · � is the euclidean norm.

Data availability
The data generated analysed during the current study are available from the corresponding author on reason-
able request.

Received: 6 June 2023; Accepted: 21 August 2023

References
1. Preskill, J. Quantum computing in the NISQ era and beyond. Quantum 2, 79. https://​doi.​org/​10.​22331/q-​2018-​08-​06-​79 (2018).
2. Bharti, K. et al. Noisy intermediate-scale quantum algorithms. Rev. Mod. Phys. 94, 015004. https://​doi.​org/​10.​1103/​RevMo​dPhys.​
94.​015004 (2022).
3. Eisert, J. et al. Quantum certification and benchmarking. Nat. Rev. Phys. 2, 382–390. https://​doi.​org/​10.​1038/​s42254-​020-​0186-4
(2020).
4. Moll, N. et al. Quantum optimization using variational algorithms on near-term quantum devices. Quantum Sci. Technol. 3, 030503.
https://​doi.​org/​10.​1088/​2058-​9565/​aab822 (2018).
5. Lubinski, T. et al. Application-oriented performance benchmarks for quantum computing. IEEE Transactions on Quantum Engi-
neering 4, 1–32. https://​doi.​org/​10.​48550/​arXiv.​2110.​03137 (2023).
6. Helsen, J., Roth, I., Onorati, E., Werner, A. H. & Eisert, J. General framework for randomized benchmarking. PRX Quantum 3,
020357. https://​doi.​org/​10.​1103/​PRXQu​antum.3.​020357 (2022).
7. Arute, F. et al. Quantum supremacy using a programmable superconducting processor. Nature 574, 505–510. https://​doi.​org/​10.​
1038/​s41586-​019-​1666-5 (2019).
8. Cross, A. W., Bishop, L. S., Sheldon, S., Nation, P. D. & Gambetta, J. M. Validating quantum computers using randomized model
circuits. Phys. Rev. A 100, 032328. https://​doi.​org/​10.​1103/​PhysR​evA.​100.​032328 (2019).
9. Alsina, D. & Latorre, J. I. Experimental test of Mermin inequalities on a five-qubit quantum computer. Phys. Rev. A 94, 012314.
https://​doi.​org/​10.​1103/​PhysR​evA.​94.​012314 (2016).
10. Swain, M., Rai, A., Behera, B. K. & Panigrahi, P. K. Experimental demonstration of the violations of Mermin’s and Svetlichny’s
inequalities for W and GHZ states. Quantum Inf. Process. 18, 218. https://​doi.​org/​10.​1007/​s11128-​019-​2331-5 (2019).
11. Huang, W.-J. et al. Mermin’s inequalities of multiple qubits with orthogonal measurements on IBM Q 53-qubit system. Quantum
Eng. 2, e45. https://​doi.​org/​10.​1002/​que2.​45 (2020).
12. González, D., de la Pradilla, D. F. & González, G. Revisiting the Experimental Test of Mermin’s Inequalities at IBMQ. Int. J. Theor.
Phys. 59, 3756–3768. https://​doi.​org/​10.​1007/​s10773-​020-​04629-4 (2020).
13. Bäumer, E., Gisin, N. & Tavakoli, A. Demonstrating the power of quantum computers, certification of highly entangled measure-
ments and scalable quantum nonlocality. npj Quantum Inf. 7, 117. https://​doi.​org/​10.​1038/​s41534-​021-​00450-x (2021).
14. Yang, B., Raymond, R., Imai, H., Chang, H. & Hiraishi, H. Testing scalable Bell inequalities for quantum graph states on IBM
quantum devices. IEEE J. Emerg. Sel. Top. Circuits Syst. 12, 638–647. https://​doi.​org/​10.​1109/​JETCAS.​2022.​32017​30 (2022).
15. Wang, Y., Li, Y., Yin, Z.-Q. & Zeng, B. 16-qubit IBM universal quantum computer can be fully entangled. npj Quantum Inf. 4, 46.
https://​doi.​org/​10.​1038/​s41534-​018-​0095-x (2018).
16. Mooney, G. J., White, G. A. L., Hill, C. D. & Hollenberg, L. C. L. Whole-device entanglement in a 65-qubit superconducting
quantum computer. Adv. Quantum Technol. 4, 2100061. https://​doi.​org/​10.​1002/​qute.​20210​0061 (2021).
17. Wei, K. X. et al. Verifying multipartite entangled Greenberger–Horne–Zeilinger states via multiple quantum coherences. Phys.
Rev. A 101, 032343. https://​doi.​org/​10.​1103/​PhysR​evA.​101.​032343 (2020).
18. Mooney, G. J., White, G. A. L., Hill, C. D. & Hollenberg, L. C. L. Generation and verification of 27-qubit Greenberger–Horne–
Zeilinger states in a superconducting quantum computer. J. Phys. Commun. 5, 095004. https://​doi.​org/​10.​1088/​2399-​6528/​ac1df7
(2021).
19. Hoban, M. J., Campbell, E. T., Loukopoulos, K. & Browne, D. E. Non-adaptive measurement-based quantum computation and
multi-party Bell inequalities. New J. Phys. 13, 023014. https://​doi.​org/​10.​1088/​1367-​2630/​13/2/​023014 (2011).
20. Demirel, B., Weng, W., Thalacker, C., Hoban, M. & Barz, S. Correlations for computation and computation for correlations. npj
Quantum Inf. 7, 29. https://​doi.​org/​10.​1038/​s41534-​020-​00354-2 (2021).

Scientific Reports | (2023) 13:15428 | https://doi.org/10.1038/s41598-023-41025-4 8

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Vol:.(1234567890)
www.nature.com/scientificreports/

21. Maciejewski, F. B., Zimborás, Z. & Oszmaniec, M. Mitigation of readout noise in near-term quantum devices by classical post-
processing based on detector tomography. Quantum 4, 257. https://​doi.​org/​10.​22331/q-​2020-​04-​24-​257 (2020).
22. Qiskit contributors. Qiskit: An open-source framework for quantum computing. https://​doi.​org/​10.​5281/​zenodo.​25735​05 (2023).
23. Mackeprang, J., Bhatti, D., Hoban, M. J. & Barz, S. The power of qutrits for non-adaptive measurement-based quantum computing.
New J. Phys. 25, 073007. https://​doi.​org/​10.​1088/​1367-​2630/​acdf77 (2023).
24. Bell, J. S. On the Einstein Podolsky Rosen paradox. Phys. Phys. Fizika 1, 195–200. https://​doi.​org/​10.​1103/​Physi​csPhy​sique​Fizika.​
1.​195 (1964).
25. Brunner, N., Cavalcanti, D., Pironio, S., Scarani, V. & Wehner, S. Bell nonlocality. Rev. Mod. Phys. 86, 419–478. https://​doi.​org/​10.​
1103/​RevMo​dPhys.​86.​419 (2014).
26. Werner, R. F. & Wolf, M. M. All-multipartite Bell-correlation inequalities for two dichotomic observables per site. Phys. Rev. A 64,
032112. https://​doi.​org/​10.​1103/​PhysR​evA.​64.​032112 (2001).
27. Żukowski, M. & Brukner, Č. Bell’s theorem for general N-qubit states. Phys. Rev. Lett. 88, 210401. https://​doi.​org/​10.​1103/​PhysR​
evLett.​88.​210401 (2002).
28. Raussendorf, R. Contextuality in measurement-based quantum computation. Phys. Rev. A 88, 022322. https://​doi.​org/​10.​1103/​
PhysR​evA.​88.​022322 (2013).
29. Efron, B. & Tibshirani, R. An introduction to the bootstrap. In Chapman & Hall/CRC Monographs on Statistics & Applied Prob-
ability (Taylor & Francis, 1994).
30. Gerard, B. & Kong, M. String Abstractions for Qubit Mapping. Preprint at https://​doi.​org/​10.​48550/​arXiv.​2111.​03716 (2021).
31. Piroli, L., Styliaris, G. & Cirac, J. I. Quantum circuits assisted by local operations and classical communication: Transformations
and phases of matter. Phys. Rev. Lett. 127, 220503. https://​doi.​org/​10.​1103/​PhysR​evLett.​127.​220503 (2021).

Acknowledgements
We thank Jonas Helsen for the useful suggestions and explanations, Alexandra R. van den Berg for the fruitful
discussions, Roeland Wiersema for the helpful tips and Chewon Cho (조채원) for the in-depth explanations.
We thank Lukas Rückle and Christopher Thalacker for the explanations. We acknowledge support from the Carl
Zeiss Foundation, the Centre for Integrated Quantum Science and Technology (IQST ), the Federal Ministry of
Education and Research (BMBF, projects SiSiQ and PhotonQ), the Federal Ministry for Economic Affairs and
Climate Action (BMWK, project PlanQK), and the Competence Center Quantum Computing Baden-Württem-
berg (funded by the Ministerium für Wirtschaft, Arbeit und Tourismus Baden-Württemberg, project QORA).
We acknowledge the use of IBM Quantum services for this work. The views expressed are those of the authors,
and do not reflect the official policy or position of IBM or the IBM Quantum team.

Author contributions
J.M., D.B. designed the experiments; J.M. performed the experiments and acquired the experimental data. J.M.
carried out theoretical derivations and wrote all the code; J.M., D.B., S.B. carried out theoretical calculations and
the data analysis. J.M., D.B., S.B. wrote the manuscript. S.B. supervised the project.

Funding
Open Access funding enabled and organized by Projekt DEAL.

Competing interests
The authors declare no competing interests.

Additional information
Supplementary Information The online version contains supplementary material available at https://​doi.​org/​
10.​1038/​s41598-​023-​41025-4.
Correspondence and requests for materials should be addressed to S.B.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.

© The Author(s) 2023

Scientific Reports | (2023) 13:15428 | https://doi.org/10.1038/s41598-023-41025-4 9

Content courtesy of Springer Nature, terms of use apply. Rights reserved
Vol.:(0123456789)
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

[email protected]