Year: 2023
Type: preprint
Abstract: Quantum Random Access Optimizer (QRAO) is a quantum-relaxation based optimization algorithm proposed by Fuller et al. that utilizes Quantum Random Access Code (QRAC) to encode multiple variables of binary optimization in a single qubit. The approximation ratio bound of QRAO for the maximum cut problem is $0.555$ if the bit-to-qubit compression ratio is $3$x, while it is $0.625$ if the compression ratio is $2$x, thus demonstrating a trade-off between space efficiency and approximability. In this research, we extend the quantum-relaxation by using another QRAC which encodes three classical bits into two qubits (the bit-to-qubit compression ratio is $1.5$x) and obtain its approximation ratio for the maximum cut problem as $0.722$. Also, we design a novel quantum relaxation that always guarantees a $2$x bit-to-qubit compression ratio which is unlike the original quantum relaxation of Fuller~et~al. We analyze the condition when it has a non-trivial approximation ratio bound $\left(>\frac{1}{2}\right)$. We hope that our results lead to the analysis of the quantum approximability and practical efficiency of the quantum-relaxation based approaches.
Source: arXiv (Cornell University)
Authors: Teramoto, Kosei, Raymond, Rudy, Wakakuwa, Eyuri, Imai, Hiroshi
Institutions: The University of Tokyo, Keio University, IBM Japan, Ltd. (Japan)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: Keio University
OpenAlex ID: W4321472396