Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 27 of the 27 references it cites were found in a registry or an index.

Benchmarking the performance of portfolio optimization with QAOA

Open access

Abstract

We present a detailed study of portfolio optimization using different versions of the quantum approximate optimization algorithm (QAOA). For a given list of assets, the portfolio optimization problem is formulated as quadratic binary optimization constrained on the number of assets contained in the portfolio. QAOA has been suggested as a possible candidate for solving this problem (and similar combinatorial optimization problems) more efficiently than classical computers in the case of a sufficiently large number of assets. However, the practical implementation of this algorithm requires a careful consideration of several technical issues, not all of which are discussed in the present literature. The present article intends to fill this gap and thereby provides the reader with a useful guide for applying QAOA to the portfolio optimization problem (and similar problems). In particular, we will discuss several possible choices of the variational form and of different classical algorithms for finding the corresponding optimized parameters. Viewing at the application of QAOA on error-prone NISQ hardware, we also analyse the influence of statistical sampling errors (due to a finite number of shots) and gate and readout errors (due to imperfect quantum hardware). Finally, we define a criterion for distinguishing between ‘easy’ and ‘hard’ instances of the portfolio optimization problem.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1007/s11128-022-03766-5

Cites

27 of 27 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Flow chart of the method for solving the portfolio optimization problem with QAOA as proposed in this article
Fig. 1 Flow chart of the method for solving the portfolio optimization problem with QAOA as proposed in this articleCC-BY · From the PDF
Fig. 2 a QAOAparameterlandscapeat p = 1, for an exemplary portfolio optimization instance with n = 10 assets. To enable a direct comparison with ( b ), the parameters γ, β are expressed in terms of m 1 = 2 γ and m 2 = 2 β , respectively. The colour code shows the deviation of 〈 ˆ F 〉 γ,β = 〈 ˆ F 〉 ( lin , 1 ) m 1 , m 2 from the optimal solution F min on a logarithmic scale. Three pronounced minima are visible in black. b For higher depth p = 7 with linear ansatz for the parameters ⃗ γ and ⃗ β , see Eqs. (25, 26), only one of these minima (close to the bottom left corner) survives (Color figure online)
Fig. 2 a QAOAparameterlandscapeat p = 1, for an exemplary portfolio optimization instance with n = 10 assets. To enable a direct comparison with ( b ), the parameters γ, β are expressed in terms of m 1 = 2 γ and m 2 = 2 β , respectively. The colour code shows the deviation of 〈 ˆ F 〉 γ,β = 〈 ˆ F 〉 ( lin , 1 ) m 1 , m 2 from the optimal solution F min on a logarithmic scale. Three pronounced minima are visible in black. b For higher depth p = 7 with linear ansatz for the parameters ⃗ γ and ⃗ β , see Eqs. (25, 26), only one of these minima (close to the bottom left corner) survives (Color figure online)CC-BY · From the PDF
Fig. 3 a Mean deviation 1 -r of the approximation ratio from the optimal solution as a function of the QAOA depth p for 20 randomly chosen portfolio optimization instances with n = 5 assets (and budget B = 2), using different mixers (see legend) and the statevector simulator (without measurement noise). The error bars display the standard deviation among the 20 random instances. The full XY -mixer and QAMPA (which requires less CNOT gates) show the fastest convergence towards the optimal solution. b Mean probability P of obtaining the optimal portfolio with standard deviation (error bars) as a function of QAOA depth p for the statevector simulator. c, d Sameas a, b , but using the qasm simulator including measurement noise due to a finite number of 1000 shots instead of the statevector simulator. Significant differences with respect to the statevector simulator are only visible in the vicinity of the optimal solution ( r and P close to 1)
Fig. 3 a Mean deviation 1 -r of the approximation ratio from the optimal solution as a function of the QAOA depth p for 20 randomly chosen portfolio optimization instances with n = 5 assets (and budget B = 2), using different mixers (see legend) and the statevector simulator (without measurement noise). The error bars display the standard deviation among the 20 random instances. The full XY -mixer and QAMPA (which requires less CNOT gates) show the fastest convergence towards the optimal solution. b Mean probability P of obtaining the optimal portfolio with standard deviation (error bars) as a function of QAOA depth p for the statevector simulator. c, d Sameas a, b , but using the qasm simulator including measurement noise due to a finite number of 1000 shots instead of the statevector simulator. Significant differences with respect to the statevector simulator are only visible in the vicinity of the optimal solution ( r and P close to 1)CC-BY · From the PDF
Fig. 4 Same as in Fig. 3 with ensemble of portfolio instances with n = 10 and budget B = 5. The main conclusions, in particular concerning the superiority of the full XY and the QAMPA mixer, are similar as in Fig. 3
Fig. 4 Same as in Fig. 3 with ensemble of portfolio instances with n = 10 and budget B = 5. The main conclusions, in particular concerning the superiority of the full XY and the QAMPA mixer, are similar as in Fig. 3CC-BY · From the PDF
Image 5
CC-BY · From the PDF
Fig. 6 Expectation value of the investigated QAOA instance for the standard mixer, the ring XY -mixer, the parity ring XY -mixer, the QAMPA, and the full XY -mixer (from left to right) and noise strength ˜ η ∈ { 0 , 0 . 1 , 1 , 2 , 4 , 10 } (from bottom to top), where ˜ η is the noise strength normalized by the size of the quantum computation, i.e. the number of quantum gates in the QAOA ansatz circuit
Fig. 6 Expectation value of the investigated QAOA instance for the standard mixer, the ring XY -mixer, the parity ring XY -mixer, the QAMPA, and the full XY -mixer (from left to right) and noise strength ˜ η ∈ { 0 , 0 . 1 , 1 , 2 , 4 , 10 } (from bottom to top), where ˜ η is the noise strength normalized by the size of the quantum computation, i.e. the number of quantum gates in the QAOA ansatz circuitCC-BY · From the PDF
Fig. 7 The approximation ratio of the investigated QAOA instance as a function of depolarizing noise strength η for different mixers using a density matrix simulator with 8192 measurement samples. The classical optimizer COBYLA [28] was used
Fig. 7 The approximation ratio of the investigated QAOA instance as a function of depolarizing noise strength η for different mixers using a density matrix simulator with 8192 measurement samples. The classical optimizer COBYLA [28] was usedCC-BY · From the PDF
Image 8
CC-BY · From the PDF
Image 9
CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 26 references

Research works printed but not linked: 2

  1. Markowitz, H.: Portfolio selection. J. Finance 7, 77 (1952)
  2. Nielsen, M.A., Chuang, I.L.: Quantum Computation and Quantum Information: 10th Anniversary Edition. Cambridge University Press, Cambridge (2010)

Cited by

See all 106

Metadata

Authors
Brandhofer, S., Braun, D., Dehn, V., Hellstern, G., Hüls, M., Ji, Y., Polian, I., Bhatia, A. S., & Wellens, T.
Affiliations
  • University of Stuttgart (DE)
  • University of Tübingen (DE)
  • Fraunhofer Institute for Applied Solid State Physics (DE)
  • Ravensburg University of Cooperative Education (DE)
Corresponding author
Dehn, Vanessa
Venue
Quantum Information Processing
Published
2022-12-16
Volume
22
Issue
1
ISSN
  • 1570-0755 (ISSN-L)
  • 1573-1332
OpenAlex
W4313293044
Cited by
106
References
26
Citations by year
2026: 27 · 2025: 32 · 2024: 32 · 2023: 12 · 2022: 3
FWCI
6.50
Type
journal article
Language
en
Open access
Licence
Locations
  • Quantum Information Processing (publishedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion)
  • Fraunhofer-Publica (Fraunhofer-Gesellschaft) (submittedVersion)
  • arXiv (Cornell University) (cc-by)
Funders
Ministerium für Wirtschaft, Arbeit und Wohnungsbau Baden-Württemberg (QORA)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Quantum Computing Algorithms and Architecture
  • Social Sciences › Decision Sciences › Management Science and Operations Research › Risk and Portfolio Optimization
  • Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Quantum Information and Cryptography
Keywords
portfolio optimization, quantum approximate optimization algorithm, variational formulation, parameter optimization, NISQ devices, shot noise, benchmarking

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Benchmarking the performance of portfolio optimization with QAOA. https://app.cochat.ai/papers/10.1007/s11128-022-03766-5
The paper
Brandhofer, S., Braun, D., Dehn, V., Hellstern, G., Hüls, M., Ji, Y., Polian, I., Bhatia, A. S., & Wellens, T. (2022). Benchmarking the performance of portfolio optimization with QAOA. Quantum Information Processing. https://doi.org/10.1007/s11128-022-03766-5

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.