Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A quantum system control method based on enhanced reinforcement learning

Wenjie Liu, Bosi Wang, Jihao Fan, Yebo Ge, Mohammed Zidan

TL;DR

The paper addresses high-fidelity quantum-state control under limited hardware resources. It proposes QSC-ERL, a framework that combines a Q-table with an enhanced neural network and a heuristic guidance term to accelerate learning, using a three-switch control paradigm for a spin-1/2 system. Empirical results show QSC-ERL achieves fidelities near 0.99 with substantially fewer episodes (≈42) than baseline RL methods, demonstrating superior efficiency under resource constraints. This approach offers a scalable, data-efficient route for quantum state preparation and control, with potential to generalize to broader quantum-control problems and hardware-limited scenarios.

Abstract

Traditional quantum system control methods often face different constraints, and are easy to cause both leakage and stochastic control errors under the condition of limited resources. Reinforcement learning has been proved as an efficient way to complete the quantum system control task. To learn a satisfactory control strategy under the condition of limited resources, a quantum system control method based on enhanced reinforcement learning (QSC-ERL) is proposed. The states and actions in reinforcement learning are mapped to quantum states and control operations in quantum systems. By using new enhanced neural networks, reinforcement learning can quickly achieve the maximization of long-term cumulative rewards, and a quantum state can be evolved accurately from an initial state to a target state. According to the number of candidate unitary operations, the three-switch control is used for simulation experiments. Compared with other methods, the QSC-ERL achieves close to 1 fidelity learning control of quantum systems, and takes fewer episodes to quantum state evolution under the condition of limited resources.

A quantum system control method based on enhanced reinforcement learning

TL;DR

The paper addresses high-fidelity quantum-state control under limited hardware resources. It proposes QSC-ERL, a framework that combines a Q-table with an enhanced neural network and a heuristic guidance term to accelerate learning, using a three-switch control paradigm for a spin-1/2 system. Empirical results show QSC-ERL achieves fidelities near 0.99 with substantially fewer episodes (≈42) than baseline RL methods, demonstrating superior efficiency under resource constraints. This approach offers a scalable, data-efficient route for quantum state preparation and control, with potential to generalize to broader quantum-control problems and hardware-limited scenarios.

Abstract

Traditional quantum system control methods often face different constraints, and are easy to cause both leakage and stochastic control errors under the condition of limited resources. Reinforcement learning has been proved as an efficient way to complete the quantum system control task. To learn a satisfactory control strategy under the condition of limited resources, a quantum system control method based on enhanced reinforcement learning (QSC-ERL) is proposed. The states and actions in reinforcement learning are mapped to quantum states and control operations in quantum systems. By using new enhanced neural networks, reinforcement learning can quickly achieve the maximization of long-term cumulative rewards, and a quantum state can be evolved accurately from an initial state to a target state. According to the number of candidate unitary operations, the three-switch control is used for simulation experiments. Compared with other methods, the QSC-ERL achieves close to 1 fidelity learning control of quantum systems, and takes fewer episodes to quantum state evolution under the condition of limited resources.
Paper Structure (15 sections, 15 equations, 3 figures, 2 tables, 2 algorithms)

This paper contains 15 sections, 15 equations, 3 figures, 2 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Learning control of quantum systems
  4. Quantum control landscapes
  5. Reinforcement learning
  6. Methods
  7. Problem modeling
  8. Enhanced reinforcement learning
  9. Heuristic function based on enhanced neural network
  10. The parameters updating method
  11. Simulation experiments
  12. Settings
  13. Evaluation index
  14. Results and analysis
  15. Conclusion

Figures (3)

  • Figure 1: An overview of enhanced reinforcement learning: the orange rectangle is given by the environment. $S_{t}$ and $S_{t+1}$ are input into the enhanced neural network which is abbreviated to E network. The algorithm selects $Q^*_{t}$ according to $a_{t}$ from $Q_{S_{t}}$and $maxQ_{t+1}$ from $Q_{S_{t+1}}$ respectively. Then calculating the loss for updating the enhanced neural network between "the blue rectangles".
  • Figure 2: The enhanced neural network architecture: For each state s fed into the network, the network extracts features and outputs Q values.
  • Figure 3: The comparison of fidelity between algorithms. (a)Fidelity of the TQL algorithm. (b) Fidelity of the PG algorithm. (c) Fidelity of the DQL algorithm. (d) Fidelity of the NN-QSC algorithm. (e) Fidelity of the DRL-QSC algorithm. (f) Fidelity of the QSC-ERL algorithm.