Dropout is all you need: robust two-qubit gate with reinforcement learning
arXiv:2312.06335 · doi:10.1103/PhysRevA.110.032614
Abstract
In the realm of quantum control, reinforcement learning, a prominent branch of machine learning, emerges as a competitive candidate for computer-assisted optimal design for experiments. This study investigates the extent to which guidance from human experts is necessary for the effective implementation of reinforcement learning in designing quantum control protocols. Specifically, we focus on the engineering of a robust two-qubit gate within a nuclear magnetic resonance system, utilizing a combination of analytical solutions as prior knowledge and techniques from the field of computer science. Through extensive benchmarking of different models, we identify dropout, a widely-used method for mitigating overfitting in machine learning, as an especially robust approach. Our findings demonstrate the potential of incorporating computer science concepts to propel the development of advanced quantum technologies.
arXiv comment11 pages, 4 figures
References in corpus (39)
- An Introduction to Deep Reinforcement Learning
- Quantum computation with ions in thermal motion
- Shortcuts to adiabaticity: concepts, methods, and applications
- Fast optimal frictionless atom cooling in harmonic traps
- Quantum reinforcement learning
- Lewis-Riesenfeld invariants and transitionless tracking algorithm
- Experimental Realization of Universal Geometric Quantum Gates with Solid-State Spins
- Implementation of universal quantum gates based on nonadiabatic geometric phases
- Experimental Realization of Non-Abelian Geometric Gates
- Reinforcement Learning in Different Phases of Quantum Control
- Robust quantum control by shaped pulse
- Unconventional geometric quantum computation
- Experimental fault-tolerant universal quantum gates with solid-state spins under ambient conditions
- Automated Search for new Quantum Experiments
- Arbitrarily accurate composite pulses
- Plug-and-Play Approach to Non-adiabatic Geometric Quantum Gates
- Analytically solvable driven time-dependent two-level quantum systems
- When does reinforcement learning stand out in quantum control? A comparative study on state preparation
- Deep Reinforcement Learning for Quantum Gate Control
- Learning Robust and High-Precision Quantum Controls
- Coherent Transport of Quantum States by Deep Reinforcement Learning
- Computer-inspired Quantum Experiments
- Global optimization of quantum dynamics with AlphaZero deep exploration
- Reinforcement Learning for Many-Body Ground-State Preparation Inspired by Counterdiabatic Driving
- Measurement Based Feedback Quantum Control With Deep Reinforcement Learning for Double-well Non-linear Potential
- Automatic spin-chain learning to explore the quantum speed limit
- Breaking Adiabatic Quantum Control with Deep Learning
- Dynamical invariants for quantum control of four-level systems
- High-fidelity geometric gate for silicon-based spin qubits
- Batch Optimization of Frequency-Modulated Pulses for Robust Two-qubit Gates in Ion Chains
- Classifying global state preparation via deep reinforcement learning
- Geometric quantum gates in liquid-state NMR based on a cancellation of dynamical phases
- Supervised learning for robust quantum control in composite-pulse systems
- Quantum Dropout: On and Over the Hardness of Quantum Approximate Optimization Algorithm
- Approximation of quantum control correction scheme using deep neural networks
- Deep reinforcement learning for robust quantum optimization
- Non-Adiabatic Universal Holonomic Quantum Gates Based on Abelian Holonomies
- Robust optimization for quantum reinforcement learning control using partial observations
- Machine-learning-inspired quantum optimal control of nonadiabatic geometric quantum computation via reverse engineering