Reinforcement learning to learn quantum states for Heisenberg scaling accuracy
arXiv:2412.02334 · doi:10.1002/qute.202500206
Abstract
Learning quantum states is a crucial task for realizing quantum information technology. Recently, neural approaches have emerged as promising methods for learning quantum states. We propose a meta-learning model that utilizes reinforcement learning (RL) to optimize the process of learning quantum states. To improve the data efficiency of the RL, we introduce an action repetition strategy inspired by curriculum learning. The RL agent significantly improves the sample efficiency of learning random quantum states, and achieves infidelity scaling close to the Heisenberg limit. We also show that the RL agent trained using 3-qubit states can generalize to learning up to 5-qubit states. These results highlight the utility of RL-driven meta-learning to enhance the efficiency and generalizability of learning quantum states. Our approach can be applied to improve quantum control, quantum optimization, and quantum machine learning.
arXiv comment15 pages, 8 figures
References in corpus (50)
- Hardware-efficient Variational Quantum Eigensolver for Small Molecules and Quantum Magnets
- Barren plateaus in quantum neural network training landscapes
- Quantum Circuit Learning
- Average Entropy of a Subsystem
- An introduction to quantum machine learning
- Evaluating analytic gradients on quantum hardware
- Experimental Quantum State Tomography of Optical Fields and Ultrafast Statistical Sampling
- Unknown Quantum States and Operations, a Bayesian View
- Efficient Learning for Deep Quantum Neural Networks
- Efficient quantum state tomography
- Quantum state estimation
- Transfer learning in hybrid classical-quantum neural networks
- Efficient Representation of Quantum Many-body States with Deep Neural Networks
- Reconstructing quantum states with generative models
- Quantum generalisation of feedforward neural networks
- Efficient tomography of a quantum many-body system
- Effect of barren plateaus on gradient-free optimization
- Diluted maximum-likelihood algorithm for quantum tomography
- Experimental neural network enhanced quantum tomography
- Eigenvalue statistics of the real Ginibre ensemble
- Learning Quantum Systems
- Adaptive quantum state tomography improves accuracy quadratically
- Higher Order Derivatives of Quantum Neural Networks with Barren Plateaus
- Self-guided quantum tomography
- A practical and efficient approach for Bayesian quantum state estimation
- Learning to Optimize Variational Quantum Circuits to Solve Combinatorial Problems
- Experimental Adaptive Bayesian Tomography
- Experimental Comparison of Efficient Tomography Schemes for a Six-Qubit State
- Random Bures mixed states and the distribution of their purity
- Quantum POMDPs
- Learning hard quantum distributions with variational autoencoders
- Deep Reinforcement Learning for Quantum State Preparation with Weak Nonlinear Measurements
- Attention-based Quantum Tomography
- Deep Reinforcement Learning Control of Quantum Cartpoles
- Variational Quantum Circuits for Quantum State Tomography
- Transfer learning for scalability of neural-network quantum states
- Collective vs local measurements in qubit mixed state estimation
- Self-Correcting Quantum Many-Body Control using Reinforcement Learning with Tensor Networks
- Natural Evolutionary Strategies for Variational Quantum Computation
- Gradient Ascent Pulse Engineering with Feedback
- Variational quantum process tomography
- Quantum State Tomography using Quantum Machine Learning
- Efficient Quantum State Sample Tomography with Basis-dependent Neural-networks
- Learning unknown pure quantum states
- Adaptive Quantum State Tomography with Active Learning
- Neural networks for quantum state tomography with constrained measurements
- Enhancing quantum state tomography via resource-efficient attention-based neural networks
- Efficient learning of mixed-state tomography for photonic quantum walk
- Experimental Learning of Pure Quantum States using Sequential Single-Shot Measurement Outcomes
- Neural network assisted quantum state and process tomography using limited data sets