arxiv: 2605.11626 · v1 · submitted 2026-05-12 · 💻 cs.IT · cs.ET· math.IT

Recognition: 2 theorem links

· Lean Theorem

Performance of QUBO-Formulated MIMO Detection Under Hardware Precision Constraints

Seyedkhashayar Hashemi , Elisabetta Valiante , Ignacio Rozada , Moslem Noori

Authors on Pith no claims yet

Pith reviewed 2026-05-13 01:18 UTC · model grok-4.3

classification 💻 cs.IT cs.ETmath.IT

keywords MIMO detectionQUBO formulationfinite precisionquantizationhardware accelerationbit error rateheterogeneous quantization

0 comments

The pith

Heterogeneous quantization of QUBO coefficients matches full-precision MIMO detection performance with far fewer bits.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper examines how finite hardware precision affects QUBO reformulations of MIMO detection, a step needed to run parallel physics-inspired solvers on specialized hardware. The authors derive the probability distributions of the QUBO matrix entries and then introduce both uniform and non-uniform quantization rules that draw on either instantaneous channel values or their long-term statistics. A sufficient precision bound is stated that keeps the quantized problem from changing the optimal solution. Large-scale simulations for multiple antenna counts and modulation orders up to 256-QAM show that the non-uniform rule reaches the same bit-error-rate curve as full precision while using substantially lower average bit width. The result supplies concrete hardware-design rules for choosing the quantization method that best balances accuracy and resource use.

Core claim

The paper derives closed-form probability distributions for the entries of the QUBO matrix that arises from MIMO detection and uses them to construct homogeneous and heterogeneous quantization mappings keyed to either instantaneous channel state or its statistical moments. A sufficient condition on the number of bits is proved that guarantees the quantized coefficients leave the global optimum unchanged. Extensive Monte-Carlo trials across MIMO dimensions and constellations up to 256-QAM demonstrate that the heterogeneous scheme reproduces the floating-point bit-error-rate performance while requiring markedly fewer bits than any fixed-precision alternative.

What carries the argument

Heterogeneous quantization scheme that allocates variable bit widths to QUBO coefficients according to channel-state information or its statistics, thereby controlling the impact of quantization noise on the detection optimum.

If this is right

Hardware accelerators for QUBO MIMO detection can be built with lower average word length and still achieve the floating-point error floor.
Designers obtain explicit rules for selecting homogeneous versus heterogeneous quantization once system size and modulation order are known.
The same quantization approach scales to higher-order constellations without requiring a proportional increase in bit width.
Power and area budgets for non-von-Neumann MIMO detectors shrink because fewer bits are stored and routed while detection quality is preserved.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The statistical quantization rule could be pre-computed once per deployment environment, allowing fixed-point hardware that never needs real-time channel feedback.
The derived precision bound supplies an a-priori sizing formula that could be checked analytically for new QUBO problems outside wireless communications.
If the bound remains tight under correlated fading, the method may generalize directly to massive MIMO and mmWave scenarios without additional simulation campaigns.

Load-bearing premise

The analytically derived distributions of QUBO entries and the sufficient-precision bound continue to hold once realistic hardware noise and imperfect channel estimates are introduced.

What would settle it

A hardware-in-the-loop experiment or high-fidelity noise model in which the heterogeneous quantizer at the predicted bit depths produces a measurable rise in bit-error rate relative to the full-precision reference under the same channel realizations.

Figures

Figures reproduced from arXiv: 2605.11626 by Elisabetta Valiante, Ignacio Rozada, Moslem Noori, Seyedkhashayar Hashemi.

**Figure 1.** Figure 1: Examples of M-QAM constellations with M = 16 (left) and M = 64 (right). The minimum distance between symbols is q 6 M−1 , such that the constellation’s average power is normalized to unity. II. MIMO DETECTION: SYSTEM MODEL AND PROBLEM FORMULATION A. System Model We consider an uplink communication link between the Nt transmit antennas and the Nr receive antennas. The received complex-valued signal vector y… view at source ↗

**Figure 2.** Figure 2: Probability distribution functions (PDF) of QUBO matrix entries [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: Maximum quantization error limit δmax versus precision for different MIMO systems with Eb/N0 = 10 dB. Then, we define ϵmin as the p-th percentile of the above normal distribution. For δmax, we take a numerical approach by first calculating δup from Eq. (22) on a large number of channel realizations and then setting δmax to the p-th percentile of the collected samples. In what follows, we compare the analyt… view at source ↗

**Figure 4.** Figure 4: Bit error rate for MIMO detection performed after applying preprocessing with different quantization schemes and precision. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗

**Figure 5.** Figure 5: Summary of quantization schemes’ performance to highlight important values of [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗

**Figure 6.** Figure 6: Bit error rate obtained by an exhaustive solver on MIMO systems [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗

read the original abstract

The evolution of multiple-input, multiple-output (MIMO) systems requires the efficient detection algorithms to overcome the exponential computational complexity of optimal maximum likelihood detection. Reformulating MIMO detection as a quadratic unconstrained binary optimization (QUBO) problem enables the use of highly parallel, physics-inspired, hardware-accelerated solvers and non-von Neumann architectures. However, embedding continuous-valued QUBO coefficients into hardware introduces quantization noise due to finite precision, which can severely degrade detection accuracy. This paper presents a rigorous analysis of the performance impact of finite-precision, hardware-accelerated QUBO solvers in MIMO detection. We analytically derive the probability distribution functions of the QUBO matrix entries and introduce novel homogeneous and heterogeneous quantization schemes based on either instantaneous channel state information or its statistical features. We further derive a sufficient condition on the precision required to maintain the optimal solution after quantization. Extensive numerical experiments, across various MIMO system sizes and modulation orders (up to 256-QAM), show that heterogeneous quantization matches the full-precision baseline bit error rate using significantly fewer bits than homogeneous approaches. We provide hardware-aware guidelines for selecting the optimal quantization strategy.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper derives PDFs for QUBO entries and a sufficient precision condition, then shows heterogeneous quantization based on CSI can match full-precision BER with fewer bits in simulations.

read the letter

The main thing to know is that the authors derive the probability distributions of the QUBO matrix entries for MIMO detection and prove a sufficient condition on bit precision that preserves the optimal solution after quantization. They introduce homogeneous and heterogeneous quantization schemes that adapt to instantaneous or statistical channel state information, and their simulations across MIMO sizes and up to 256-QAM show the heterogeneous version matches full-precision bit error rate while using noticeably fewer bits overall. That analytical step plus the practical bit-saving result is the clearest contribution here, and the hardware-aware guidelines at the end follow directly from it. The work is grounded in the QUBO reformulation and cites the relevant prior literature without obvious gaps in the chain of reasoning. The softer spot is that the numerical experiments isolate quantization noise but do not add other realistic hardware effects such as thermal noise, device mismatch, or time-varying errors. If those push effective precision below the derived threshold, the observed BER equivalence could break even when the bit allocation meets the sufficient condition. The derivations themselves appear independent of the simulations, so the analytical claim stands on its own, but the hardware realism gap is real and worth flagging. This paper is for people working on physics-inspired solvers or low-power MIMO hardware implementations. It has enough new analysis and scaled simulations to deserve a serious referee, though revisions should test the condition under fuller hardware noise models.

Referee Report

2 major / 2 minor

Summary. The manuscript analyzes finite-precision effects in hardware-accelerated QUBO solvers for MIMO detection. It derives the probability density functions of the QUBO matrix entries, proposes homogeneous and heterogeneous quantization schemes that use either instantaneous CSI or its statistical properties, and proves a sufficient precision condition guaranteeing that quantization preserves the optimal solution. Extensive Monte Carlo simulations across MIMO sizes and modulation orders up to 256-QAM are reported to show that the heterogeneous scheme matches full-precision bit-error-rate performance while requiring substantially fewer bits than homogeneous quantization.

Significance. If the PDF derivations and sufficient-condition proof are correct and the simulations faithfully represent the quantization model, the work supplies concrete, hardware-aware design rules for deploying QUBO-based MIMO detectors on resource-limited or physics-inspired accelerators. The combination of closed-form distributional analysis, an explicit optimality-preserving bound, and broad numerical validation constitutes a useful contribution to the intersection of information theory and hardware-constrained optimization.

major comments (2)

[Numerical Experiments] Numerical Experiments section: the reported BER curves compare heterogeneous quantization against full-precision and homogeneous baselines, yet the simulation description gives no indication that additional hardware impairments (thermal noise, device mismatch, or time-varying quantization errors) are injected on top of the modeled quantization. Because the sufficient precision condition is derived under an ideal quantization model, any unmodeled noise that reduces effective precision below the derived threshold could invalidate the observed BER equivalence; this is load-bearing for the central performance claim.
[Sufficient Condition] Sufficient Condition (derivation section): the proof that a given bit-width guarantees retention of the same optimal solution relies on the analytically derived PDFs of the QUBO entries. Without the explicit steps showing how the tail bounds or concentration inequalities are applied to the matrix entries, it is impossible to confirm that the condition remains valid once the QUBO formulation is embedded in a realistic MIMO channel with estimation error.

minor comments (2)

[Abstract] The abstract states that guidelines for selecting the quantization strategy are provided, but these guidelines are not summarized in a table or clearly enumerated subsection; adding such a summary would improve readability.
[Quantization Schemes] Notation for the heterogeneous quantization thresholds (e.g., how the statistical features of the channel are mapped to bit allocations) should be introduced earlier and used consistently throughout the derivations.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the thorough review and insightful comments on our manuscript. We address each of the major comments below and outline the revisions we plan to make.

read point-by-point responses

Referee: [Numerical Experiments] Numerical Experiments section: the reported BER curves compare heterogeneous quantization against full-precision and homogeneous baselines, yet the simulation description gives no indication that additional hardware impairments (thermal noise, device mismatch, or time-varying quantization errors) are injected on top of the modeled quantization. Because the sufficient precision condition is derived under an ideal quantization model, any unmodeled noise that reduces effective precision below the derived threshold could invalidate the observed BER equivalence; this is load-bearing for the central performance claim.

Authors: We appreciate the referee highlighting this aspect. Our simulations are intentionally focused on evaluating the impact of the proposed quantization schemes under the ideal quantization model used in the derivation of the sufficient precision condition. The goal is to demonstrate that heterogeneous quantization can achieve equivalent BER performance to full precision with reduced bit-widths, without additional impairments. We agree that in practical hardware deployments, other noise sources may be present and could necessitate adjustments to the precision requirements. To address this, we will revise the Numerical Experiments section to explicitly state that only quantization effects are modeled and include a discussion noting that real-world hardware impairments represent an important consideration for future extensions of this work. revision: partial
Referee: [Sufficient Condition] Sufficient Condition (derivation section): the proof that a given bit-width guarantees retention of the same optimal solution relies on the analytically derived PDFs of the QUBO entries. Without the explicit steps showing how the tail bounds or concentration inequalities are applied to the matrix entries, it is impossible to confirm that the condition remains valid once the QUBO formulation is embedded in a realistic MIMO channel with estimation error.

Authors: Thank you for this comment. The derivation of the sufficient condition in the manuscript uses the closed-form PDFs of the QUBO matrix entries (derived in Section III) and applies union bounds on the probability that quantization perturbs the optimal solution. We will expand the proof in the revised manuscript to include the detailed steps, explicitly showing the application of tail bounds and concentration inequalities to the individual matrix entries. Regarding channel estimation error, the current analysis assumes perfect CSI, which is standard for deriving fundamental limits in MIMO detection studies. With imperfect CSI, the QUBO coefficients would be perturbed differently, requiring a separate analysis. We will add a remark in the manuscript clarifying this assumption and suggesting it as a direction for future research. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivations are independent analytical steps

full rationale

The paper analytically derives the PDFs of QUBO matrix entries and a sufficient precision condition to preserve the optimal solution, presenting these as first-principles results based on the QUBO formulation and quantization noise model. Numerical experiments then validate that heterogeneous quantization matches full-precision BER with fewer bits. No step reduces by construction to a fitted parameter renamed as prediction, a self-citation chain, or an ansatz smuggled via prior work; the central claims remain self-contained against external benchmarks like full-precision baselines. This matches the default expectation for non-circular papers.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Based solely on the abstract, no explicit free parameters, ad-hoc axioms, or invented entities are identifiable; the work relies on standard MIMO channel models and QUBO reformulation assumptions from prior literature.

axioms (1)

domain assumption Standard statistical models for MIMO channels and QUBO coefficient distributions hold under finite precision
Invoked implicitly when deriving PDFs and quantization effects

pith-pipeline@v0.9.0 · 5510 in / 1226 out tokens · 34330 ms · 2026-05-13T01:18:59.969464+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We analytically derive the probability distribution functions of the QUBO matrix entries and ... derive a sufficient condition on the precision required to maintain the optimal solution after quantization.
IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Extensive numerical experiments ... heterogeneous quantization matches the full-precision baseline bit error rate

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

26 extracted references · 26 canonical work pages · 1 internal anchor

[1]

Verd ´u,Multiuser Detection

S. Verd ´u,Multiuser Detection. Cambridge, UK: Cambridge University Press, 1998

work page 1998
[2]

Paulraj, R

A. Paulraj, R. Nabar, and D. Gore,Introduction to Space-Time Wireless Communications. Cambridge, UK: Cambridge University Press, 2003

work page 2003
[3]

A Universal Lattice Code Decoder for Fading Channels,

E. Viterbo and J. Boutros, “A Universal Lattice Code Decoder for Fading Channels,”IEEE Transactions on Information Theory, vol. 45, no. 5, pp. 1639–1642, 1999

work page 1999
[4]

Layered Space-Time Architecture for Wireless Com- munication in a Fading Environment When Using Multi-Element An- tennas,

G. J. Foschini, “Layered Space-Time Architecture for Wireless Com- munication in a Fading Environment When Using Multi-Element An- tennas,”Bell Labs Technical Journal, vol. 1, no. 2, pp. 41–59, 1996

work page 1996
[5]

DeepSIC: Deep Soft Interference Cancellation for Multiuser MIMO Detection,

N. Shlezinger, R. Fu, and Y . C. Eldar, “DeepSIC: Deep Soft Interference Cancellation for Multiuser MIMO Detection,”IEEE Transactions on Wireless Communications, vol. 20, no. 2, pp. 1349–1362, 2020. 9 4 8 12 16 20 24 Nt = Nr = 4, nb M = 4 M = 16 M = 64 M = 256 4 8 12 16 20 24 Nt = Nr = 16, nb Hom Hom–Stat Het Het–Stat 4 8 12 16 20 24 Nt = Nr = 32, nb Hom ...

work page 2020
[6]

Soft-Input Soft-Output Single Tree-Search Sphere Decoding,

C. Studer and H. B ¨olcskei, “Soft-Input Soft-Output Single Tree-Search Sphere Decoding,”IEEE Transactions on Information Theory, vol. 56, no. 10, pp. 4827–4842, 2010

work page 2010
[7]

MMSE- Based Lattice-Reduction for Near-ML Detection of MIMO Systems,

D. W ¨ubben, R. B ¨ohnke, V . K ¨uhn, and K.-D. Kammeyer, “MMSE- Based Lattice-Reduction for Near-ML Detection of MIMO Systems,” in ITG Workshop on Smart Antennas (IEEE Cat. No.04EX802), Munich, Germany, 2004

work page 2004
[8]

Leveraging quantum annealing for large MIMO processing in centralized radio access networks,

M. Kim, D. Venturelli, and K. Jamieson, “Leveraging quantum annealing for large MIMO processing in centralized radio access networks,” in Proceedings of the ACM Special Interest Group on Data Communica- tion, ser. SIGCOMM ’19. New York, USA: Association for Computing Machinery, 2019, pp. 241–255

work page 2019
[9]

High-Order modulation large MIMO detector based on physics-inspired methods,

Q.-G. Zeng, X. Cui, X.-Z. Tao, J.-Q. Hu, S.-J. Pan, W. E. I. Sha, and M.-H. Yung, “High-Order modulation large MIMO detector based on physics-inspired methods,”Physical Review Applied, vol. 24, p. 034027, 2025

work page 2025
[10]

Physics-inspired heuristics for soft MIMO detection in 5G new radio and beyond,

M. Kim, S. Mandr `a, D. Venturelli, and K. Jamieson, “Physics-inspired heuristics for soft MIMO detection in 5G new radio and beyond,” in Proceedings of the 27th Annual International Conference on Mobile Computing and Networking, ser. MobiCom ’21. New York, USA: Association for Computing Machinery, 2021, pp. 42–55

work page 2021
[11]

Ising Machines’ Dynamics and Regularization for Near-Optimal MIMO De- tection,

A. K. Singh, K. Jamieson, P. L. McMahon, and D. Venturelli, “Ising Machines’ Dynamics and Regularization for Near-Optimal MIMO De- tection,”IEEE Transactions on Wireless Communications, vol. 21, no. 12, pp. 11 080–11 094, 2022

work page 2022
[12]

Probabilistic Computers for MIMO Detection: From Sparsification to 2D Parallel Tempering,

M. M. H. Sajeeb, C. Delacour, K. Callahan-Coray, S. Seshan, T. Srimani, and K. Y . Camsari, “Probabilistic Computers for MIMO Detection: From Sparsification to 2D Parallel Tempering,” 2026, arXiv:2601.09037

work page arXiv 2026
[13]

A fully parallel densely connected probabilistic Ising machine with inertia for real-time applications

R. Zhu, A. K. Singh, J. Laydevant, F. O. Wu, A. Kapelyan, D. Venturelli, K. Jamieson, and P. L. McMahon, “A fully parallel densely connected probabilistic Ising machine with inertia for real-time applications,” 2026, arXiv:2604.17109

work page internal anchor Pith review Pith/arXiv arXiv 2026
[14]

What is the resolution of the control parameters in D-Wave devices?

D-Wave Systems, “What is the resolution of the control parameters in D-Wave devices?” Published: May 7, 2019, Accessed: March 3, 2026, https://support.dwavesys.com/hc/en-us/community/posts/ 360029507594-What-is-the-resolution-of-the-control-parameters-in-D- Wave-devices. 10

work page 2019
[15]

100,000-spin coherent Ising machine,

T. Honjo, T. Sonobe, K. Inaba, T. Inagaki, T. Ikuta, Y . Yamada, T. Kazama, K. Enbutsu, T. Umeki, R. Kasahara, K. Kawarabayashi, and H. Takesue, “100,000-spin coherent Ising machine,”Science Advances, vol. 7, no. 40, p. eabh0952, 2021

work page 2021
[16]

Mismatch Characterization and Calibration for Accurate and Automated Analog Design,

S. Shapero and P. Hasler, “Mismatch Characterization and Calibration for Accurate and Automated Analog Design,”IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 60, no. 3, pp. 548–556, 2013

work page 2013
[17]

An ultra-low energy internally analog, externally digital vector-matrix multiplier based on NOR flash memory technology,

M. R. Mahmoodi and D. Strukov, “An ultra-low energy internally analog, externally digital vector-matrix multiplier based on NOR flash memory technology,” inProceedings of the 55th Annual Design Au- tomation Conference, ser. DAC ’18. New York, USA: Association for Computing Machinery, 2018, pp. 1–6

work page 2018
[18]

Thousands of conductance levels in memristors integrated on CMOS,

M. Rao, H. Tang, J. Wu, W. Song, M. Zhang, W. Yin, Y . Zhuo, F. Kiani, B. Chen, X. Jiang, H. Liu, H. Chen, R. Midya, F. Ye, H. Jiang, Z. Wang, M. Wu, M. Hu, H. Wang, Q. Xia, N. Ge, J. Li, and J. Yang, “Thousands of conductance levels in memristors integrated on CMOS,”Nature, vol. 615, pp. 823–829, 2023

work page 2023
[19]

Multi Digit Ising Mapping for Low Precision Ising Solvers,

A. K. Singh and K. Jamieson, “Multi Digit Ising Mapping for Low Precision Ising Solvers,” 2024, arXiv:2404.05631

work page arXiv 2024
[20]

Versatile stochastic dot product circuits based on nonvolatile memories for high perfor- mance neurocomputing and neurooptimization,

M. Mahmoodi, M. Prezioso, and D. Strukov, “Versatile stochastic dot product circuits based on nonvolatile memories for high perfor- mance neurocomputing and neurooptimization,”Nature Communica- tions, vol. 10, p. 5113, 2019

work page 2019
[21]

Power-efficient combinatorial optimization using intrinsic noise in memristor Hopfield neural networks,

F. Cai, S. Kumar, T. Van Vaerenbergh, X. Sheng, R. Liu, C. Li, Z. Liu, M. Foltin, S. Yu, Q. Xia, J. J. Yang, R. Beausoleil, W. D. Lu, and J. P. Strachan, “Power-efficient combinatorial optimization using intrinsic noise in memristor Hopfield neural networks,”Nature Electronics, vol. 3, pp. 409–418, 2020

work page 2020
[22]

Memristor-based hardware and algorithms for higher-order Hopfield optimization solver outperforming quadratic Ising machines,

M. Hizzani, A. Heittmann, G. Hutchinson, D. Dobrynin, T. V . Vaeren- bergh, T. Bhattacharya, A. Renaudineau, D. Strukov, and J. P. Strachan, “Memristor-based hardware and algorithms for higher-order Hopfield optimization solver outperforming quadratic Ising machines,” in2024 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE, 2024, pp. 1–5

work page 2024
[23]

Computing high-degree polynomial gradients in memory,

T. Bhattacharya, G. H. Hutchinson, G. Pedretti, X. Sheng, J. Ignowski, T. Van Vaerenbergh, R. Beausoleil, J. P. Strachan, and D. B. Strukov, “Computing high-degree polynomial gradients in memory,”Nature Com- munications, vol. 15, p. 8211, 2024

work page 2024
[24]

PySA: Fast Simulated Annealing in Na- tive Python,

S. Mandr `a, A. Akbari Asanjan, L. Brady, A. Lott, D. E. Bernal Neira, and H. Munoz Bauza, “PySA: Fast Simulated Annealing in Na- tive Python,” Released: March 12, 2023, Accessed: May 6, 2026, https://github.com/nasa/PySA

work page 2023
[25]

Nonlinear optimization algorithm using mono- tonically increasing quantization resolution,

J. Seok and J.-S. Kim, “Nonlinear optimization algorithm using mono- tonically increasing quantization resolution,”ETRI Journal, vol. 45, no. 1, pp. 119–130, 2023

work page 2023
[26]

MMGaP: Multi-User MIMO Detection and Precoding using GPU-assisted Physics-inspired Computation,

A. K. Singh and K. Jamieson, “MMGaP: Multi-User MIMO Detection and Precoding using GPU-assisted Physics-inspired Computation,” 2025, arXiv:2510.01579. APPENDIXA Proof of Theorem 1:Expression (9) gives us the diagonal entries of the QUBO matrixQ, wheni=j: Qi,i = NrX k=1 c∗ kWk,i +c kW ∗ k,i +|W k,i|2 = 2 NrX k=1 ℜ(c∗ kWk,i) +|W k,i|2 = 2ℜ " NrX k=1 Wk,i c∗...

work page arXiv 2025