基于Kraus参数化信念强化学习的连续量子反馈控制
Continuous Quantum Feedback Control via Kraus-Parameterized Belief Reinforcement Learning
浏览论文内容
中文总结 AI 辅助
该研究提出Kraus参数化信念强化学习方法，将量子态几何嵌入学习循环，在模拟量子比特上实现稳定反馈控制，提升了控制稳定性，为可靠量子反馈控制提供了实用归纳偏置。
中文摘要 AI 辅助
量子反馈控制需要对带噪声的连续测量记录进行操作，而无法直接获取底层量子态。我们提出Kraus参数化信念强化学习（Kraus-Parameterized Belief Reinforcement Learning），该流程中，受施蒂费尔流形（Stiefel manifold）约束的循环编码器会生成密度矩阵估计值，这些估计值从构造上保证为半正定且迹归一化，将量子态几何直接嵌入学习循环中。随后，近端策略优化（PPO）算法的策略网络会将这些物理上有效的信念状态映射为连续控制动作。在模拟的连续监测量子比特上，该策略实现了稳定的反馈控制，维持测量条件下的信念保真度约为0.77-0.80，且在标称条件和分布外条件下，与参数匹配的LSTM历史基线相比，回报方差显著更低。尽管原始目标保真度的提升较为有限，但几何约束保证了物理上有效、可解释的信念表示，并在测量效率低下和动态突变情况下表现出明显更稳定的控制效果。这些结果表明，物理启发的神经记忆是实现可靠量子反馈控制的实用归纳偏置。
英文摘要
Quantum feedback control requires acting on noisy continuous measurement records without direct access to the underlying quantum state. We propose Kraus-Parameterized Belief Reinforcement Learning, a pipeline in which a recurrent encoder, constrained to the Stiefel manifold, produces density-matrix estimates that are guaranteed positive-semidefinite and trace-normalized by construction, embedding quantum state geometry directly into the learning loop. A Proximal Policy Optimization (PPO) actor then maps these physically valid belief states to continuous control actions. On a simulated continuously monitored qubit, the resulting policy achieves stable feedback control, maintaining a measurement-conditioned belief fidelity of approximately 0.77-0.80 and exhibiting substantially lower return variance than a parameter-matched LSTM-history baseline across both nominal and out-of-distribution conditions. Although gains in raw target fidelity are modest, the geometric constraint guarantees a physically valid, interpretable belief representation and yields markedly more stable control under measurement inefficiency and abrupt dynamics switches. These results indicate that physics-informed neural memory is a practical inductive bias for reliable quantum feedback control.
发表机构
- SRM Institute of Science and Technology（SRM科学技术学院）
- Fractal AI Research（分形AI研究院）
机构由 AI 辅助整理，请以论文原文为准。
