Thermometry of one-dimensional Bose gases with neural networks

Abstract

We design a neural network to extract and process features from absorption images taken of one-dimensional Bose gases in the quasi-condensate regime. Specifically, the network is trained to predict both the temperature of single realizations of the system and the uncertainty thereof. For multiple realizations, the individual predictions can be combined in an estimate of the mean temperature, improving precision. We benchmark our model on both simulated and experimentally measured data and compare it to the established method of density ripples thermometry. We find the predictions of the two methods compatible, although the neural network reaches similar precision needing much fewer realizations, thus highlighting the efficiency gain achievable when incorporating neural networks into analysis of data from cold gas experiments. Further, we study feature maps to reveal which local features of the condensate are extracted by the network and how said features correlate with properties of the system. A similar analysis could be employed to uncover physical relations in more complex systems.

arXiv comment18 pages, 9 figures

References in corpus (40)

• Adam: A Method for Stochastic Optimization
• Many-Body Physics with Ultracold Gases
• The density-matrix renormalization group
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Atomic Scattering in Presence of an External Confinement and a Gas of Impenetrable Bosons
• Machine learning phases of matter
• Relaxation and Pre-thermalization in an Isolated Quantum System
• Transport in Out-of-Equilibrium XXZ Chains: Exact Profiles of Charges and Currents
• Learning phase transitions by confusion
• Experimental Observation of a Generalized Gibbs Ensemble
• Emergent hydrodynamics in integrable quantum systems out of equilibrium
• Atom Chips
• Efficient Representation of Quantum Many-body States with Deep Neural Networks
• Extension of Bogoliubov theory to quasi-condensates
• On solving quantum many-body problems by experiment
• Identifying Quantum Phase Transitions using Artificial Neural Networks on Experimental Data
• Observations of density fluctuations in an elongated Bose gas: ideal gas and quasi-condensate regimes
• Fast machine-learning online optimization of ultra-cold-atom experiments
• Sub-Poissonian fluctuations in a 1D Bose gas: from the quantum quasi-condensate to the strongly interacting regime
• Machine Learning in Electronic Quantum Matter Imaging Experiments
• Linear response theory for a pair of coupled one-dimensional condensates of interacting atoms
• Integrating Neural Networks with a Quantum Simulator for State Reconstruction
• Probing three-body correlations in a quantum gas using the measurement of the third moment of density fluctuations
• Single-particle-sensitive imaging of freely propagating ultracold atoms
• Classifying Snapshots of the Doped Hubbard Model with Machine Learning
• Two-point density correlations of quasicondensates in free expansion
• Extension of the Generalized Hydrodynamics to the Dimensional Crossover Regime
• Making, probing and understanding Bose-Einstein condensates
• Designing Arbitrary One-dimensional Potentials on an Atom Chip
• Fluctuations and stochastic processes in one-dimensional many-body quantum systems
• Decay and recurrence of non-Gaussian correlations in a quantum many-body system
• Machine learning enhanced global optimization by clustering local environments to enable bundled atomic energies
• Applying machine learning optimization methods to the production of a quantum gas
• Monitoring squeezed collective modes of a one-dimensional Bose gas after an interaction quench using density ripples analysis
• Single-exposure absorption imaging of ultracold atoms using deep learning
• Deep Learning-Assisted Classification of Site-Resolved Quantum Gas Microscope Images
• Correlations and dynamics of tunnel-coupled one-dimensional Bose gases
• From observations to complexity of quantum states via unsupervised learning
• Density ripples in expanding low-dimensional gases as a probe of correlations
• Microscopic atom optics: from wires to an atom chip

Cited by in corpus (9)

• Emergent Pauli blocking in a weakly interacting Bose gas
• Correlation properties of a one-dimensional repulsive Bose gas at finite temperature
• The Whitham approach to Generalized Hydrodynamics
• Sample-efficient estimation of entanglement entropy through supervised learning
• Thermal fading of the $1/k^4$-tail of the momentum distribution induced by the hole anomaly
• Identifying diffusive length scales in one-dimensional Bose gases
• Measurement of total phase fluctuation in cold-atomic quantum simulators
• Systematic analysis of relative phase extraction in one-dimensional Bose gases interferometry
• Particle-hole origin of thermal beating in dipole-compression modes of a 1D Bose gas