Variational Quantum Eigensolver with Constraints (VQEC): Solving Constrained Optimization Problems via VQE

Thinh Viet Le,Vassilis Kekatos

2024-04-27

Abstract:Variational quantum approaches have shown great promise in finding near-optimal solutions to computationally challenging tasks. Nonetheless, enforcing constraints in a disciplined fashion has been largely unexplored. To address this gap, this work proposes a hybrid quantum-classical algorithmic paradigm termed VQEC that extends the celebrated VQE to handle optimization with constraints. As with the standard VQE, the vector of optimization variables is captured by the state of a variational quantum circuit (VQC). To deal with constraints, VQEC optimizes a Lagrangian function classically over both the VQC parameters as well as the dual variables associated with constraints. To comply with the quantum setup, variables are updated via a perturbed primal-dual method leveraging the parameter shift rule. Among a wide gamut of potential applications, we showcase how VQEC can approximately solve quadratically-constrained binary optimization (QCBO) problems, find stochastic binary policies satisfying quadratic constraints on the average and in probability, and solve large-scale linear programs (LP) over the probability simplex. Under an assumption on the error for the VQC to approximate an arbitrary probability mass function (PMF), we provide bounds on the optimality gap attained by a VQC. Numerical tests on a quantum simulator investigate the effect of various parameters and corroborate that VQEC can generate high-quality solutions.

Quantum Physics,Machine Learning,Optimization and Control

What problem does this paper attempt to address?

The paper attempts to address the problem of how to effectively handle optimization problems with constraints in Variational Quantum Algorithms (VQA). Specifically, existing variational quantum methods (such as the Variational Quantum Eigensolver, VQE) perform well in handling unconstrained optimization problems but have shortcomings when dealing with constrained optimization problems. This paper proposes a new hybrid quantum-classical algorithm framework—Constrained Variational Quantum Eigensolver (VQEC)—to extend the capabilities of VQE, enabling it to handle optimization problems with constraints.

### Main Issues
1. **Limitations of Existing Methods**:
- **VQE and Other VQA Methods**: While these methods perform well in handling unconstrained optimization problems, they face challenges when dealing with constrained optimization problems.
- **Handling Constraints**: Existing methods typically handle constraints by converting them into penalty terms, but this requires careful selection of penalty parameters, which may lead to numerical issues.

2. **Need for New Methods**:
- **Systematic Handling of Constraints**: There is a need for a systematic approach to handle optimization problems with constraints, rather than simply using penalty terms.
- **Efficiency and Accuracy**: The new method needs to maintain computational efficiency while ensuring the accuracy of the optimization results.

### Solution
The VQEC framework proposed in this paper addresses the above issues in the following ways:
1. **Lagrangian Dual Method**: Utilizes the Lagrangian dual method to convert the original problem into a dual problem, thereby systematically handling constraints.
2. **Perturbed Primal-Dual Method**: Employs the Perturbed Primal-Dual (PPD) method to update the parameters of the Variational Quantum Circuit (VQC) and the Lagrange multipliers, ensuring the convergence of the algorithm.
3. **Combination of Quantum and Classical Computing**: Measures observables using a quantum computer and updates parameters using a classical computer, achieving efficient hybrid computation.

### Application Examples
VQEC can be applied to various optimization problems, including but not limited to:
1. **Quadratic Constrained Binary Optimization (QCBO)**: Handles quadratic binary optimization problems with constraints.
2. **Stochastic Policy Design**: Designs stochastic binary policies that meet average or probabilistic constraints.
3. **Large-Scale Linear Programming (LP)**: Solves large-scale linear programming problems on the probability simplex.

### Summary
By proposing the VQEC framework, this paper fills the gap in Variational Quantum Algorithms for handling optimization problems with constraints, providing new tools and methods for the further development of quantum computing in practical applications.