Existence of a robust optimal control process for efficient measurements in a two-qubit system

Basic Information

• Paper ID: 2506.19122
• Title: Existence of a robust optimal control process for efficient measurements in a two-qubit system
• Authors: Ricardo Rodriguez (McPherson College), Nam Nguyen (Boeing Research & Technology), Elizabeth Behrman (Wichita State University), Andy C. Y. Li (Fermi National Accelerator Laboratory), James Steck (Wichita State University)
• Classification: quant-ph (Quantum Physics)
• Publication Date: November 24, 2025 (arXiv v3)
• Paper Link: https://arxiv.org/abs/2506.19122v3

Abstract

Verification of quantum entanglement is crucial for quality control in quantum communication. This paper proposes an efficient protocol for directly verifying two-qubit entanglement of known target states through single expectation value measurements. The method employs the concurrence metric to provide precise entanglement quantification without requiring quantum state tomography. The authors prove the existence of a unitary transformation capable of driving the initial state of a two-qubit system to a specified final state, where the trace of the initial state's concurrence is directly given by measurement of a selected observable. Furthermore, the authors implement an optimal control process for this transformation and demonstrate its effectiveness through numerical simulations. The study also shows that the process exhibits robustness against environmental noise. This method offers advantages in directly verifying entanglement at low circuit depth and is applicable to industrial-scale entanglement generation quality control.

Research Background and Motivation

1. Research Problem

This paper addresses the efficient verification of two-qubit entangled states in quantum communication. Specifically, it aims to precisely quantify the degree of entanglement through the minimum number of measurements when the entanglement level is known.

2. Problem Significance

• Quantum Communication Foundation: Quantum entanglement is a core resource for quantum communication and quantum computing, with applications in quantum key distribution, quantum secure direct communication, and quantum secret sharing.
• Industrial Requirements: Practical applications of future quantum internet and quantum communication require industrial-scale entanglement system production and quality control.
• Resource Optimization: Verification of two-qubit entanglement generated by production protocols plays a critical role in quality control.

3. Limitations of Existing Methods

• Quantum State Tomography: While a direct method for measuring entanglement, it requires measurement of complete state information, resulting in low efficiency.
• Approximate Methods: Existing approximate entanglement measures and methods utilizing quantum storage involve trade-offs in accuracy or resource consumption.
• Lack of Theoretical Guarantees: Previous computational experiments (the authors' 2008 work) lacked rigorous mathematical proofs.

4. Research Motivation

The authors envisioned a practical application scenario: particle pair sources with known entanglement levels used for specific applications require verification through sampling sufficient particle pairs, driving them to a final state using a known control process, where entanglement is verified through single expectation value measurements. If the initial state possesses different entanglement values, the control process will produce values different from expected. This is more efficient than full tomography and exhibits robustness to constant parameter drift.

Core Contributions

1. Theoretical Proof: Rigorously proves the existence of a unitary transformation capable of driving any two-qubit initial state to a specified final state, such that measurement of the σ_z⊗σ_z observable directly yields the concurrence of the initial state.
2. Controllability Theory: Proves that the two-qubit system employing a QMA-complete Hamiltonian (Equation 2) possesses density matrix controllability (DMC), with dynamic Lie algebra su(4).
3. Robustness Guarantee: Proves that the control system maintains DMC properties even in the presence of Hamiltonian drift (Corollary II.1.1), providing theoretical assurance for practical applications.
4. Optimal Control Algorithm: Designs and implements an optimal control algorithm based on variational methods, employing a forward-backward gradient process in GRAPE/Krotov style.
5. Numerical Verification: Through MATLAB/Octave implementation, validates the algorithm's effectiveness on 100-1000 different density matrix samples with relative error controlled within 5%.
6. Mathematical Justification: Provides rigorous mathematical foundation for the authors' early 2008 computational experiments.

Detailed Methodology

Task Definition

Input: Initial density matrix ρ(0) of a two-qubit system with known concurrence C
Output:

1. Final state density matrix ρ(t_f) satisfying Tr((σ_z⊗σ_z)·ρ(t_f)) = C
2. Control path u(t) = {κ_i(t), ε_i(t), ζ(t), ν(t)} implementing unitary evolution from ρ(0) to ρ(t_f)
   Constraints:

• ρ(t_f) must be unitarily equivalent to ρ(0) (same eigenvalue spectrum)
• Evolution obeys the Liouville-Von Neumann equation
• Control cost minimization

Model Architecture

1. Hamiltonian Design

The paper employs a QMA-complete Hamiltonian:

H = κ_A σ_x⊗1 + κ_B 1⊗σ_x + ε_A σ_z⊗1 + ε_B 1⊗σ_z + ζ σ_z⊗σ_z + ν σ_x⊗σ_x

Where:

• κ_i: tunneling amplitudes
• ε_i: biases
• ζ, ν: inter-qubit couplings

2. Controllability Proof Framework

Dynamic Lie Algebra (DLA) Method:

• Base operator set S = {iσ_x⊗1, i1⊗σ_x, iσ_z⊗1, i1⊗σ_z, iσ_z⊗σ_z, iσ_x⊗σ_x}
• Generate DLA through Lie bracket (commutator) calculations
• First round produces 7 new linearly independent operators
• Second round produces 2 additional linearly independent operators
• Total of 15 linearly independent operators = dim(su(4))
• Conclusion: DLA = su(4), system possesses DMC

3. Final State Existence Proof

Key Theorem (Theorem II.3):

• Define unitary equivalence set: O_λ = {Uρ(0)U† | U ∈ U(4)}
• Define measurement function: f(ρ) = Tr((σ_z⊗σ_z)·ρ)
• Apply intermediate value theorem: f is continuous on compact connected set O_λ, must attain all values between its minimum m and maximum M
• Calculate extrema:
  • M = (λ_1 + λ_2) - (λ_3 + λ_4)
  • m = (λ_3 + λ_4) - (λ_1 + λ_2)
• From concurrence definition: C = max{0, λ_1 - λ_2 - λ_3 - λ_4}
• Proof: m ≤ 0 ≤ C ≤ M, therefore ∃ρ(t_f) ∈ O_λ such that f(ρ(t_f)) = C

4. Optimal Control Design

Cost Functional (Equation 4):

J = 1/2[d - ⟨O(t_f)⟩]² + ∫Tr(λ†(1/(iℏ)[H,ρ] - ∂ρ/∂t))dt + 1/8∫Tr(H†H)dt

Contains three components:

• Measurement error term: minimizes deviation between concurrence and measured value
• Dynamic constraint: Lagrange multiplier ensures compliance with Liouville-Von Neumann equation
• Control cost: minimizes number of interventions

Euler-Lagrange Equations:

1. Control equations: ∂H/∂u = 0, yielding 6 equations (corresponding to 6 control parameters)
2. Adjoint equations: ∂λ/∂t = 1/(iℏ)H,λ
3. Boundary conditions: λ_f = -(d - ϕ(ρ_f))(σ_z⊗σ_z)

Two-Point Boundary Value Problem:

• ρ(t) = U(t)ρ_i U†(t), with initial condition ρ(0) known
• λ(t) = U†(t)λ_f U(t), with terminal condition λ_f determined by Equation 8

Technical Innovations

1. Single-Shot Measurement Verification

Compared to traditional quantum state tomography, this method requires measurement of only a single observable σ_z⊗σ_z expectation value, dramatically reducing measurement count.

2. Exact Quantification

Unlike approximate methods, this approach provides exact concurrence values without approximation error.

3. Theoretical Completeness

First to provide complete mathematical proof chain for this class of control problems:

• Controllability existence (Theorem II.1)
• Robustness guarantee (Corollary II.1.1)
• Final state existence (Theorem II.3)

4. Robust Control Design

Proves that even with Hamiltonian drift H_d, the system maintains su(4) controllability, meaning any unwanted drift can be actively compensated in real-time.

5. Discretization Algorithm

Discretizes the continuous-time problem into N time slices with constant Hamiltonian within each slice, employing "bang-bang" style control for convenient numerical implementation.

Experimental Setup

Dataset

• Generation Method: Density matrices generated using parameterization method described by Kong and Ting 33
• Sampling Strategy: Uniform sampling in parameter space
• Sample Scale: Each group contains 100-1000 different density matrices
• State Types: Includes both pure and mixed states

Evaluation Metrics

Relative Error:

Relative Error = |C(ρ(0)) - Tr(σ_z⊗σ_z ρ(t_f))| / C(ρ(0))

• Target threshold: 5%
• Measures deviation between measured value and true concurrence

Implementation Details

• Software Platform: MATLAB/Octave
• Time Steps: N = 4 (for most density matrices)
• Learning Rate: η (gradient descent update step size)
• Convergence Criterion: ∑l ∑k {1/(iℏ)Tr(λ{k+1}∂H/∂u_k, ρ_k) + u{l,k}} < ε
• Initialization: Control variables randomly initialized or using prior knowledge
• Optimization Method: Gradient descent with update rule u_{k,new} = u_{k,old} - η∂H/∂u_k

Algorithm Flow

1. Initialize all control variables {u_k; k=1,...,N-1}
2. Forward propagation: compute state variables ρ_k (starting from ρ_0)
3. Backward propagation: compute adjoint variables λ_k (starting from λ_N)
4. Check stopping criterion: ∂H/∂u_k ≈ 0
5. If not converged, update control variables using gradient descent
6. Repeat steps 2-5 until convergence

Experimental Results

Main Results

1. Concurrence vs. Measured Value Comparison (Figure 1)

• Data: Scatter plot showing relationship between initial state concurrence and final state σ_z⊗σ_z measurement value
• Ideal Case: Red line represents x=y function
• Observation: Data points cluster tightly around red line, indicating algorithm successfully drives final state to target location

2. Relative Error Distribution (Figure 2)

• Statistical Properties: Histogram shows relative error distribution
• Precision: Most samples have relative error within 5%
• Reliability: Algorithm stably finds valid solutions for different initial states meeting precision requirements

3. Numerical Performance

• Success Rate: Algorithm finds valid solutions for all 100-1000 samples
• Runtime: Achieves 5% relative error threshold within reasonable computation time
• Circuit Depth: Requires only 4 time steps (4 different Hamiltonians), low circuit depth

Key Findings

1. Unitary Matrix Construction

From control variables {u_k} of each time slice, one can construct unitary matrices representing the control process for that time interval. Sequential multiplication of these matrices equals the total unitary matrix U responsible for transformation between initial and final states.

2. Measurement Range for Pure and Mixed States

• Pure States (Corollary II.3.1): λ={1,0,0,0}, measurement range -1,1
• Rank-2 Mixed States (Corollary II.3.2): λ={λ_1,λ_2,0,0}, measurement range also -1,1
• Significance: For these states, different entanglement levels can be completely distinguished through single measurement

3. Theory-Practice Consistency

Numerical results validate theoretical predictions:

• Existence theorem guarantees solution existence
• Algorithm effectively finds these solutions
• Robustness manifests in numerical experiments

Related Work

1. Entanglement Verification Methods

• Pallister et al. 14: Optimal entangled state verification using local measurements
• Goswami et al. 15: Universal two-qubit entanglement detection using only two copies
• Wang and Hayashi 16: Optimal verification of two-qubit pure states
• This Paper's Advantage: Provides exact concurrence values rather than merely determining whether entanglement exists

2. Approximate Measurement Methods

• Riccardi et al. 17: Exploring relationship between fidelity and entanglement
• Riera-Sàbat et al. 18: Non-destructive verification through fidelity witnesses
• This Paper's Advantage: No approximation, direct exact measurement

3. Quantum Storage-Assisted Methods

• Chen et al. 19: Entangled state verification assisted by quantum storage
• This Paper's Advantage: No additional quantum storage resources required

4. Quantum Control Theory

• D'Alessandro 24: Foundational theory of quantum control and dynamics
• Koch et al. 30: Python implementation of Krotov method
• Khaneja et al. 31: GRAPE algorithm
• This Paper's Contribution: Applies these methods to specific entanglement verification problem

5. Authors' Prior Work

• Behrman et al. 23 (2008): Quantum algorithm design using dynamic learning
• This Paper's Contribution: Provides rigorous mathematical proof for 2008 computational experiments

Conclusions and Discussion

Main Conclusions

1. Theoretical Completeness: Proves that two-qubit systems can be driven from arbitrary initial states to final states through specific Hamiltonian (Equation 2), such that single σ_z⊗σ_z measurement directly yields concurrence.
2. Practicality: Proposed protocol is applicable to industrial-scale entanglement generation quality control with low circuit depth advantage.
3. Robustness: System exhibits inherent robustness to constant parameter drift (Corollary II.1.1), suitable for practical applications.
4. Efficiency: Compared to full tomography, requires only single measurement, dramatically reducing measurement resource consumption.
5. Exactness: Provides exact concurrence values without approximation error.

Limitations

1. System Scale Limitation

• Current Scope: Applicable only to two-qubit systems
• Generalization Challenge: While authors mention framework extensibility to N-qubit systems, specific implementation remains undemonstrated

2. Known State Assumption

• Application Scenario: Requires prior knowledge of initial state entanglement level
• Limitation: Inapplicable to completely unknown state entanglement measurement

3. Numerical Precision

• Error Threshold: 5% relative error represents compromise between computational cost and precision
• Improvement Space: Higher precision may require longer runtime

4. Experimental Implementation

• Theory vs. Practice: Paper primarily provides theoretical proof and numerical simulation
• Missing: No experimental verification on real quantum hardware

5. Control Complexity

• Parameter Count: Requires controlling 6 time-varying parameters
• Implementation Difficulty: Precise implementation of these controls on actual quantum devices may be challenging

Future Directions

1. Generalization to Multi-Qubit Systems

Authors explicitly state: "While our results here apply only to two-qubit systems, much of the framework can be generalized to N-qubit systems, and even to Z⊗N or general measurements."

2. Verification of Other Quantities

"Our results can be generalized to measurement or verification of other quantities."

3. Closed-Form Solutions

Authors mention: "While quantum dynamics is not control-affine in the traditional linear optimal control sense, one might assume a linear relationship between λ and ρ, achieving closed-form solutions for u by combining equations (i) and (ii). We leave this for future papers."

4. Hardware Experiments

Verify protocol effectiveness and robustness on real quantum devices.

5. Quantum Transfer Learning

The authors' quantum transfer learning work 35 provides some support for broader applicability.

In-Depth Evaluation

Strengths

1. Theoretical Rigor (★★★★★)

• Complete Proof Chain: From controllability to final state existence to robustness, logically rigorous
• Mathematical Tools: Employs advanced mathematical tools including Lie algebras, unitary equivalence theory, and intermediate value theorem
• Theorem-Corollary Structure: Well-organized hierarchy with rigorous proof for each conclusion

2. Novelty (★★★★☆)

• Single-Shot Measurement: Simplifies entanglement verification to measurement of single observable, important innovation
• Exact Quantification: Unlike approximate methods, provides exact concurrence values
• Theoretical Justification: Provides mathematical foundation for 17-year-old computational experiments

3. Practical Value (★★★★☆)

• Low Circuit Depth: Requires only 4 time steps, suitable for NISQ-era quantum devices
• Industrial Application: Explicitly targets industrial-scale quality control scenarios
• Robustness: Robustness to parameter drift enhances practicality

4. Methodological Contribution (★★★★☆)

• Variational Method: Combines optimal control theory with quantum control
• Gradient Algorithm: GRAPE/Krotov-style implementation provides actionable algorithm framework
• Two-Point Boundary Value Problem: Cleverly transforms control problem into standard mathematical problem

5. Writing Quality (★★★★★)

• Clear Structure: Flows logically from theory to algorithm to experiments
• Mathematical Expression: Detailed formula derivation with comprehensive appendices
• Figure Quality: Figures 1 and 2 intuitively demonstrate algorithm performance

Weaknesses

1. Insufficient Experimental Verification (★★☆☆☆)

• Numerical Simulation Only: Lacks experiments on real quantum hardware
• Simple Noise Model: Only considers Hamiltonian drift, not decoherence or measurement errors
• Environmental Noise: Claims robustness to environmental noise but provides no detailed noise model or testing

2. Insufficient Scalability Analysis (★★★☆☆)

• Limited to Two-Qubits: While generalization mentioned, no concrete scheme provided
• Computational Complexity: No analysis of computational cost scaling with qubit number
• Control Resources: No discussion of control parameter count in multi-qubit cases

3. Missing Comparative Experiments (★★☆☆☆)

• No Baseline Comparison: No direct comparison with existing entanglement verification methods (14-19)
• Efficiency Analysis: Claims higher efficiency but lacks quantitative resource consumption comparison
• Precision Comparison: No systematic comparison with approximate methods' precision

4. Incomplete Algorithm Details (★★★☆☆)

• Hyperparameter Selection: Learning rate η and convergence threshold ε lack guidance
• Initialization Strategy: Specific distribution for random initialization not specified
• Convergence Analysis: No theoretical convergence guarantee or empirical analysis provided

5. Application Scenario Limitations (★★★☆☆)

• Known State Assumption: Requires prior knowledge of entanglement level, limiting application scope
• Quality Control Specific: Inapplicable to unknown state entanglement measurement
• Specific Hamiltonian Dependency: Relies on specific control Hamiltonian form

Impact Assessment

1. Theoretical Contribution (High)

• Provides complete mathematical framework for entanglement verification in quantum control theory
• Demonstrates DMC application in entanglement measurement
• May inspire efficient measurement methods for other quantum properties

2. Practical Value (Medium-High)

• Short-term: Provides theoretical foundation for quantum communication quality control
• Medium-term: Potential application in quantum network node verification
• Long-term: If generalized to multi-qubit, impact will increase significantly

3. Reproducibility (Medium)

• Theory: Clear proofs, easy to verify
• Algorithm Implementation: Framework provided but lacks complete code
• Numerical Results: Parameter settings insufficiently detailed for complete reproduction

4. Academic Impact

• Citation Potential: Theoretical rigor may attract citations from quantum control and quantum information fields
• Follow-up Research: Opens directions for multi-qubit generalization, other quantity verification, etc.
• Interdisciplinary: Connects quantum physics, control theory, and optimization algorithms

Applicable Scenarios

1. Ideal Scenarios

• Quantum Communication Quality Control: Regular verification of known entanglement sources
• Quantum Network Nodes: Verification of distributed entangled pairs
• Quantum Device Testing: Evaluating entanglement generation protocol performance

2. Inapplicable Scenarios

• Unknown State Entanglement Detection: When expected entanglement level unknown
• Large-Scale Quantum Systems: Currently limited to two-qubit systems
• High-Noise Environments: Robustness may be insufficient under strong decoherence

3. Potential Extension Scenarios

• Quantum Sensing: Verifying entanglement resources in sensors
• Quantum Computing: Verifying entanglement produced by quantum gate operations
• Quantum Simulation: Monitoring entanglement evolution during simulation

In-Depth Technical Analysis

1. Hamiltonian Selection Rationale

The paper's chosen Hamiltonian (Equation 2) is QMA-complete, meaning:

• Universality: Theoretically capable of simulating any quantum system
• Physical Realizability: Included terms (σ_x, σ_z and couplings) are implementable on superconducting qubits, ion traps, and other platforms
• Minimality: Although containing 6 control parameters, Theorem IV.1 indicates theoretically only 2 universal Hamiltonians needed

2. Cost Functional Design

The three-term balance reflects optimization objectives:

• Measurement Error Term: Ensures reaching target concurrence
• Dynamic Constraint: Guarantees physical realizability
• Control Cost Term: 1/8 coefficient is empirically chosen, affecting control "smoothness"

3. Numerical Stability

• Matrix Exponential Computation: U(t) = exp-iH(t-t_0)/ℏ requires high precision
• Gradient Computation: Involves trace operations on density matrices and Lagrange multipliers, requiring attention to numerical error accumulation
• Convergence: No theoretical guarantee provided; may converge slowly on certain initial states

Key References

1. 21 W. K. Wootters (1998): Original concurrence definition, core metric in this paper
2. 24 D. D'Alessandro (2022): Standard textbook on quantum control theory, DMC theoretical foundation
3. 23 Behrman et al. (2008): Authors' prior work, this paper provides mathematical proof
4. 14-19: Recent entanglement verification work, representing state-of-the-art in field
5. 30-32: GRAPE/Krotov algorithms, theoretical foundation for numerical implementation

Overall Assessment

This is a paper with rigorous theory and innovative methodology. Main strengths lie in providing a complete mathematical framework for entanglement verification, proving feasibility and robustness of single-shot measurement verification. Theoretical contributions are significant, particularly DMC proof and final state existence theorem.

Main weaknesses are weak experimental verification, with only numerical simulation lacking real quantum hardware testing, and missing comparative experiments, making it difficult to assess practical advantages over existing methods. Scalability is also problematic, currently limited to two-qubit systems.

Recommended for: Quantum control theory researchers, quantum communication engineers, scholars interested in quantum entanglement measurement.

Academic Value: ★★★★☆ (4/5)
Practical Value: ★★★☆☆ (3/5)
Novelty: ★★★★☆ (4/5)
Rigor: ★★★★★ (5/5)