Andrew Zhao,Nicholas C. Rubin
Abstract:Quadratic programming over orthogonal matrices encompasses a broad class of hard optimization problems that do not have an efficient quantum representation. Such problems are instances of the little noncommutative Grothendieck problem (LNCG), a generalization of binary quadratic programs to continuous, noncommutative variables. In this work, we establish a natural embedding for this class of LNCG problems onto a fermionic Hamiltonian, thereby enabling the study of this classical problem with the tools of quantum information. This embedding is accomplished by a new representation of orthogonal matrices as fermionic quantum states, which we achieve through the well-known double covering of the orthogonal group. Correspondingly, the embedded LNCG Hamiltonian is a two-body fermion model. Determining extremal states of this Hamiltonian provides an outer approximation to the original problem, a quantum analogue to classical semidefinite relaxations. In particular, when optimizing over the \emph{special} orthogonal group our quantum relaxation obeys additional, powerful constraints based on the convex hull of rotation matrices. The classical size of this convex-hull representation is exponential in matrix dimension, whereas our quantum representation requires only a linear number of qubits. Finally, to project the relaxed solution back into the feasible space, we propose rounding procedures which return orthogonal matrices from appropriate measurements of the quantum state. Through numerical experiments we provide evidence that this rounded quantum relaxation can produce high-quality approximations.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to expand the processing ability of a class of optimization problems with rotation matrices as decision variables in quantum optimization. Specifically, this type of problem involves quadratic programming on orthogonal matrices, which encompasses a wide range of optimization problems that are difficult to represent effectively in quantum terms. These problems are an instance of the Little Non - commutative Grothendieck Problem (LNCG), that is, a generalization from binary quadratic programming to continuous, non - commutative variables. The main contribution of the paper lies in proposing a natural method to embed this type of LNCG problem into a fermionic Hamiltonian, thus enabling the use of quantum information tools to study this classical problem. Through this method, the paper not only provides a new perspective for solving these optimization problems but also explores how to use quantum computers to obtain more accurate solutions.
### Specific Methods for Solving Problems in the Paper
1. **Fermionic Embedding**: The paper proposes a new representation method, treating orthogonal matrices as fermionic quantum states, which is achieved through the known double - covering of the orthogonal group. Correspondingly, the embedded LNCG Hamiltonian is a two - body fermionic model.
2. **Quantum Relaxation**: Determining the extreme - value states of this Hamiltonian provides an outer approximation of the original problem, similar to the classical semidefinite relaxation. In particular, when optimizing on the special orthogonal group, the proposed quantum relaxation obeys strong constraint conditions based on the convex hull of rotation matrices.
3. **Quantum Rounding Algorithm**: In order to project the relaxed solution back to the feasible space, the paper proposes a rounding procedure that returns an orthogonal matrix by appropriately measuring the quantum state. Through numerical experiments, the authors provide evidence that this rounded quantum relaxation can produce high - quality approximate solutions.
### Application Background
- **Group Synchronization Problem**: It has applications in fields such as structural biology, robotics, and wireless networks. For example, in cryo - electron microscopy technology, it is necessary to estimate the best fit of molecular structures in different unknown directions.
- **Generalized Orthogonal Procrustes Problem**: It has applications in fields such as shape and image recognition, sensory analysis, and market research. The goal is to simultaneously fit multiple point clouds, allowing each point cloud to undergo an arbitrary orthogonal transformation to best align the individual points.
### Main Results
- **Quantum Hamiltonian Relaxation**: By embedding the problem into a quantum Hamiltonian, the paper provides a quantum relaxation framework that is applicable to optimization on orthogonal matrices or special orthogonal matrices.
- **Quantum Rounding Protocol**: Two rounding protocols are proposed, one based on the edge - marginal density matrix and the other based on the vertex - marginal density matrix. Both methods can effectively recover orthogonal matrices from quantum states.
### Conclusion
By embedding a class of complex optimization problems into a quantum system, the paper not only provides new methods for solving these problems but also lays the foundation for exploring the potential of quantum computing in the field of optimization. The effectiveness of these methods has been verified through numerical experiments, demonstrating the potential advantages of quantum computing in dealing with this type of problem.