An efficient and scalable variational quantum circuits approach for deep reinforcement learning

• Published in: Quantum information processing Vol. 22; no. 8
• Main Authors: Bar, Niyazi Furkan, Yetis, Hasan, Karakose, Mehmet
• Format: Journal Article
• Language: English
• Published: New York Springer US 02-08-2023
• Subjects: Data Structures and Information Theory; Mathematical Physics; Physics; Physics and Astronomy; Quantum Computing; Quantum Information Technology; Quantum Physics; Spintronics; Deep reinforcement learning; Quantum computing; Quantum machine learning; Variational quantum circuit
• ISSN: 1573-1332; 1573-1332
• DOI: 10.1007/s11128-023-04051-9

Nowadays, machine learning techniques are successfully applied to many problems in industrial and academic fields with classical computers. With the introduction of quantum simulators, the idea of using quantum-computing speed to solve these problems has become widespread. Many researchers are experimenting with using quantum circuits in various machine-learning methods to solve different problems. Due to the limited number of qubits, experiments are on simpler problems. In this study, a variational quantum circuit (VQC) was proposed using amplitude encoding to overcome the limited qubit number barrier and use the advantages of quantum computing more efficiently. The proposed amplitude encoding method and VQC were explained. Generalized by exemplifying how they can be applied to different problems. The proposed approach was applied to a navigation problem. The performance of the proposed approach was evaluated with the number of parameters, the number of qubits needed, and the success rate. As a result, the performance of the proposed approach has been verified.