Reinforcement Learning to Learn Quantum States for Heisenberg Scaling Accuracy

Learning quantum states is a crucial task for realizing quantum information technology. Recently, neural approaches have emerged as promising methods for learning quantum states. This work proposes a meta‐learning model that utilizes reinforcement learning (RL) to optimize the process of learning quantum states. To enhance the data efficiency of the RL, an action repetition strategy inspired by curriculum learning is introduced. The RL agent significantly improves the sample efficiency of learning random quantum states, and achieves infidelity scaling close to the Heisenberg limit. Also, the RL agent trained on three‐qubit states demonstrates generalization capabilities to learning up to five‐qubit states. These results highlight the utility of RL‐driven meta‐learning to enhance the efficiency and generalizability of learning quantum states. This approach can be applied to improve quantum control, quantum optimization, and quantum machine learning.