AI RESEARCH
Scalable Quantum Reinforcement Learning on NISQ Devices with Dynamic-Circuit Qubit Reuse and Grover Optimization
arXiv CS.LG
•
ArXi:2509.16002v2 Announce Type: replace-cross A scalable and resource-efficient quantum reinforcement learning framework is presented that eliminates the linear qubit-scaling barrier in multi-step quantum Marko decision processes (QMDPs). The proposed framework integrates a QMDP formulation, dynamic-circuit execution, and Grover-based amplitude amplification into a unified quantum-native architecture.