Fast machine-learning online optimization of ultra-cold-atom experiments
arXiv:1507.04964 · doi:10.1038/srep25890
Abstract
Machine-designed control of complex devices or experiments can discover strategies superior to those developed via simplified models. We describe an online optimization algorithm based on Gaussian processes and apply it to optimization of the production of Bose-Einstein condensates (BEC). BEC is typically created with an exponential evaporation ramp that is approximately optimal for s-wave, ergodic dynamics with two-body interactions and no other loss rates, but likely sub-optimal for many real experiments. Machine learning using a Gaussian process, in contrast, develops a statistical model of the relationship between the parameters it controls and the quality of the BEC produced. This is an online process, and an active one, as the Gaussian process model updates on the basis of each subsequent experiment and proposes a new set of parameters as a result. We demonstrate that the Gaussian process machine learner is able to discover a ramp that produces high quality BECs in 10 times fewer iterations than a previously used online optimization technique. Furthermore, we show the internal model developed can be used to determine which parameters are essential in BEC creation and which are unimportant, providing insight into the optimization process.
arXiv comment6 pages, 4 figures
References in corpus (17)
- Many-Body Physics with Ultracold Gases
- Ultracold atoms out of equilibrium
- Optimal quantum control using randomized benchmarking
- Coherent control using adaptive learning algorithms
- Quantum Control Landscapes
- Adaptive hybrid optimal quantum control for imprecisely characterized systems
- Are there traps in quantum control landscapes?
- Atom lasers: production, properties and prospects for precision inertial measurement
- Control Landscapes for Observable Preparation with Open Quantum Systems
- Characterization of the Critical Submanifolds in Quantum Ensemble Control Landscapes
- Bosenova and three-body loss in a Rb-85 Bose-Einstein condensate
- A scripted control system for autonomous hardware timed experiments
- Optimizing the efficiency of evaporative cooling in optical dipole traps
- A Bose-condensed, simultaneous dual species Mach-Zehnder atom interferometer
- Stochastic optimization of a cold atom experiment using a genetic algorithm
- Comment on "Are There Traps in Quantum Control Landscapes?"
- Evolutionary optimization of an experimental apparatus
Cited by in corpus (89)
- Machine learning and the physical sciences
- Autonomous discovery in the chemical sciences part I: Progress
- Reinforcement Learning in Different Phases of Quantum Control
- Identifying Quantum Phase Transitions using Artificial Neural Networks on Experimental Data
- Bayesian optimization for tuning and selecting hybrid-density functionals
- Intelligent metaphotonics empowered by machine learning
- Gryffin: An algorithm for Bayesian optimization of categorical variables informed by expert knowledge
- Quantum Machine Learning for Chemistry and Physics
- Designing nanostructures for interfacial phonon transport via Bayesian optimization
- Neutral Atom Quantum Computing Hardware: Performance and End-User Perspective
- Neural Network-based Classification of Crystal Symmetries from X-Ray Diffraction Patterns
- Deep learning enhanced Rydberg multifrequency microwave recognition
- A tweezer array with 6100 highly coherent atomic qubits
- Bayesian optimization with known experimental and design constraints for chemistry applications
- Learning in Quantum Control: High-Dimensional Global Optimization for Noisy Quantum Dynamics
- Multiparameter optimisation of a magneto-optical trap using deep learning
- Geometric Speed Limit of Accessible Many-Body State Preparation
- Unsupervised machine learning of topological phase transitions from experimental data
- Remote optimization of an ultra-cold atoms experiment by experts and citizen scientists
- Using Recurrent Neural Networks to Optimize Dynamical Decoupling for Quantum Memory
- Glassy Phase of Optimal Quantum Control
- Olympus: a benchmarking framework for noisy optimization and experiment planning
- Machine Learning for Achieving Bose-Einstein Condensation of Thulium Atoms
- How deep is deep enough? -- Quantifying class separability in the hidden layers of deep neural networks
- Experimental demonstration of shaken lattice interferometry
- Quantum state control of a Bose-Einstein condensate in an optical lattice
- Optimal control of mirror pulses for atom interferometry
- Machine learning \& artificial intelligence in the quantum domain
- Visualizing Strange Metallic Correlations in the 2D Fermi-Hubbard Model with AI
- Applying machine learning optimization methods to the production of a quantum gas
- Machine-learning-accelerated Bose-Einstein condensation
- Optimal quantum control with poor statistics
- Approaching the adiabatic timescale with machine-learning
- An atom interferometer with a shaken optical lattice
- Predicting quantum dynamical cost landscapes with deep learning
- Variational quantum algorithm with information sharing
- An unsupervised deep learning algorithm for single-site reconstruction in quantum gas microscopes
- Single-exposure absorption imaging of ultracold atoms using deep learning
- Single-site-resolved imaging of ultracold atoms in a triangular optical lattice
- Autonomous Calibration of Single Spin Qubit Operations
- Phase diagram and optimal control for n-tupling discrete time crystal
- Adaptive Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation
- Broken symmetry in a two-qubit quantum control landscape
- Demonstration of weighted graph optimization on a Rydberg atom array using local light-shifts
- High-efficiency cold-atom transport into a waveguide trap
- Perspective on new implementations of atomtronic circuits
- Preparation of ordered states in ultra-cold gases using Bayesian optimization
- Efficient Learning of Accurate Surrogates for Simulations of Complex Systems
- Integration of spectator qubits into quantum computer architectures for hardware tuneup and calibration
- Machine learner optimization of optical nanofiber-based dipole traps for cold Rb atoms
- Active Learning Algorithm for Computational Physics
- Persistent currents in ultracold gases
- The Raspberry Pi Auto-aligner: Machine Learning for Automated Alignment of Laser Beams
- Optimal amplification of the dynamical Casimir effect in a parametrically driven system
- Excitonic Wave Function Reconstruction from Near-Field Spectra Using Machine Learning Techniques
- Active Learning Approach to Optimization of Experimental Control
- High-performance Raman memory with spatio-temporal reversal
- High-field solution state DNP using cross-correlations
- Machine learning applied to quantum synchronization-assisted probing
- Bose-Einstein-condensate source on an optical-grating-based atom chip for quantum sensor applications
- Thermometry of one-dimensional Bose gases with neural networks
- Growing Extended Laughlin States in a Quantum Gas Microscope: A Patchwork Construction
- Optimizing persistent currents in a ring-shaped Bose-Einstein condensate using machine learning
- Machine Learning for Estimation and Control of Quantum Systems
- Vortex detection in atomic Bose-Einstein condensates using neural networks trained on synthetic images
- High-dimensional reinforcement learning for optimization and control of ultracold quantum gases
- Quantum Reservoir Computing for Speckle-Disorder Potentials
- Creation and manipulation of quantized vortices in Bose-Einstein condensates using reinforcement learning
- Generation of Bose-Einstein Condensates' Ground State Through Machine Learning
- Loading a quantum gas from a hybrid dimple trap to a shell trap
- Review and experimental benchmarking of machine learning algorithms for efficient optimization of cold atom experiments
- Framework for Learning and Control in the Classical and Quantum Domains
- Bayesian Optimization for Robust State Preparation in Quantum Many-Body Systems
- Combining machine learning with physics: A framework for tracking and sorting multiple dark solitons
- Deterministic quantum state transfer of atoms in a random magnetic field
- Gaussian processes for choosing laser parameters for driven, dissipative Rydberg aggregates
- Pulse shaping in the terahertz frequency range for the control of photo-excited carriers in graphene
- Bayesian optimization for state engineering of quantum gases
- Bose-Einstein condensation by polarization gradient laser cooling
- Inspiration from machine learning on example of optimization of the Bose-Einstein condensate of thulium atoms in a 1064-nm trap
- Gaussian Process Regression for Absorption Spectra Analysis of Molecular Dimers
- Fundamental Limits of Feedback Cooling Ultracold Atomic Gases
- Precise wavefunction engineering with magnetic resonance
- Fingerprint Recognition of Partial Discharge Signals in Deep Learning Enhanced Rydberg Atomic Sensors
- Bayesian optimization of non-classical optomechanical correlations
- Efficient production of sodium Bose-Einstein condensates in a hybrid trap
- Evaporative cooling by pulse width modulation (PWM) of optical dipole traps
- Multimode feedback cooling of the collective modes of a Bose-Einstein condensate
- A Robust Strontium Tweezer Apparatus for Quantum Computing