Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2310.03267 · Springer proceedings in physics

Machine Learning Detection of Correlations in Snapshots of Ultracold Atoms in Optical Lattices

Stephanie L. Striegel, Eduardo Ibarra-García-Padilla, Ehsan Khatami

Published Jan 1, 2026• 0 citations
arXivHTMLPDF

Abstract

Abstract Recent proposals have suggested the use of supervised learning with convolutional neural networks to shed light on some of the less well known phases of the Fermi-Hubbard model through the classification of snapshots from the quantum gas microscopy of ultracold atoms in optical lattices. However, there have been challenges in the interpretability of networks with more than one convolutional filter coupled to the input images. Here, we expand on previous work by considering multiple filters in the first convolutional layer and developing a process for analyzing the physical relevance of patterns obtained in the trained filters. We benchmark our approach at half-filling, where strong antiferromagnetic correlations are known to be present, and we find that upon hole doping, previously unknown patterns arise at temperatures below the tunneling amplitude. These patterns may be a signature of interesting arrangements of fermions in the lattice.

DOI
10.1007/978-981-92-0844-9_19

Related papers

Visualizing strange metallic correlations in the two-dimensional Fermi-Hubbard model with artificial intelligence

Ehsan Khatami, Elmer Guardado-Sanchez, Benjamin M. Spar +3 more

Strongly correlated phases of matter are often described in terms of straightforward electronic patterns. This has so far been the basis for studying the Fermi-Hubbard model realized with ultracold atoms. Here, we show that artificial intelligence (AI) can provide an unbiased alternative to this paradigm for phases with subtle, or even unknown, patterns. Long- and short-range spin correlations spontaneously emerge in filters of a convolutional neural network trained on snapshots of single atomic species. In the less well-understood strange metallic phase of the model, we find that a more complex network trained on snapshots of local moments produces an effective order parameter for the non-Fermi-liquid behavior. Our technique can be employed to characterize correlations unique to other phases with no obvious order parameters or signatures in projective measurements, and has implications for science discovery through AI beyond strongly correlated systems.

arXiv:2002.12310•Physical Review A•Sep 17, 2020•35 citations

Correlator convolutional neural networks as an interpretable architecture for image-like quantum matter data

Cole Miles, Annabelle Bohrdt, Ruihan Wu +7 more

Image-like data from quantum systems promises to offer greater insight into the physics of correlated quantum matter. However, the traditional framework of condensed matter physics lacks principled approaches for analyzing such data. Machine learning models are a powerful theoretical tool for analyzing image-like data including many-body snapshots from quantum simulators. Recently, they have successfully distinguished between simulated snapshots that are indistinguishable from one and two point correlation functions. Thus far, the complexity of these models has inhibited new physical insights from such approaches. Here, we develop a set of nonlinearities for use in a neural network architecture that discovers features in the data which are directly interpretable in terms of physical observables. Applied to simulated snapshots produced by two candidate theories approximating the doped Fermi-Hubbard model, we uncover that the key distinguishing features are fourth-order spin-charge correlators. Our approach lends itself well to the construction of simple, versatile, end-to-end interpretable architectures, thus paving the way for new physical insights from machine learning studies of experimental and numerical data.

arXiv:2011.03474•Nature Communications•Jun 23, 2021•50 citations

Observing a topological phase transition with deep neural networks from experimental images of ultracold atoms

Entong Zhao, Ting Hin Mak, Chengdong He +4 more

Although classifying topological quantum phases have attracted great interests, the absence of local order parameter generically makes it challenging to detect a topological phase transition from experimental data. Recent advances in machine learning algorithms enable physicists to analyze experimental data with unprecedented high sensitivities, and identify quantum phases even in the presence of unavoidable noises. Here, we report a successful identification of topological phase transitions using a deep convolutional neural network trained with low signal-to-noise-ratio (SNR) experimental data obtained in a symmetry-protected topological system of spin-orbit-coupled fermions. We apply the trained network to unseen data to map out a whole phase diagram, which predicts the positions of the two topological phase transitions that are consistent with the results obtained by using the conventional method on higher SNR data. By visualizing the filters and post-convolutional results of the convolutional layer, we further find that the CNN uses the same information to make the classification in the system as the conventional analysis, namely spin imbalance, but with an advantage concerning SNR. Our work highlights the potential of machine learning techniques to be used in various quantum systems.

arXiv:2209.10060•Optics Express•Sep 21, 2022•8 citations

Optical lattice experiments at unobserved conditions with generative adversarial deep learning

Corneel Casert, Kyle Mills, Tom Vieijra +2 more

Optical lattice experiments with ultracold atoms allow for the experimental realization of contemporary problems in many-body physics. Yet, devising models that faithfully describe experimental observables is often difficult and problem dependent; there is currently no theoretical method which accounts for all experimental observations. Leveraging the large data volume and presence of strong correlations, machine learning provides a novel avenue for the study of such systems. It has recently been proven successful in analyzing properties of experimental data of ultracold quantum gases. Here we show that generative deep learning succeeds in the challenging task of modeling such an experimental data distribution. Our method is able to produce synthetic experimental snapshots of a doped two-dimensional Fermi-Hubbard model that are indistinguishable from previously reported experimental realizations. We demonstrate how our generative model interprets physical conditions such as temperature at the level of individual configurations. We use our approach to predict snapshots at conditions and scales which are currently experimentally inaccessible, mapping the large-scale behavior of optical lattices at unseen conditions.

arXiv:2002.07055•Physical Review Research•Sep 20, 2021•13 citations

String patterns in the doped Hubbard model

Christie S. Chiu, Geoffrey Ji, Annabelle Bohrdt +6 more

Understanding strongly correlated quantum many-body states is one of the most difficult challenges in modern physics. For example, there remain fundamental open questions on the phase diagram of the Hubbard model, which describes strongly correlated electrons in solids. In this work, we realize the Hubbard Hamiltonian and search for specific patterns within the individual images of many realizations of strongly correlated ultracold fermions in an optical lattice. Upon doping a cold-atom antiferromagnet, we find consistency with geometric strings, entities that may explain the relationship between hole motion and spin order, in both pattern-based and conventional observables. Our results demonstrate the potential for pattern recognition to provide key insights into cold-atom quantum many-body systems.

arXiv:1810.03584•Science•Jul 18, 2019•184 citations

Quantum microscopy of strongly correlated phases in the Fermi-Hubbard model

Timon Hilker

Ultracold neutral atoms in optical lattices have become a powerful platform for the simulation of complex quantum systems. With our quantum gas microscope, we measure snapshots of the quantum many-body wavefunction by fully resolving the position and spin of each atom. I will present the application of this method to two quantum phases: The Haldane spin-1 phase is the prototype of a symmetry-protected topological phase, whose spin order is hidden in conventional two-point correlation functions. On our images, we can evaluate the characteristic string-correlator of this state and directly see the associated edge states of the system. In an antiferromagnetic background, holes get dressed by a surrounding cloud of ferromagnetic correlations. We study these magnetic polarons, which are a building block of the intricate physics of cuprates, by mapping out the spatial structure of the spin environment of individual holes. Upon increased doping, we observe the transition from this unusual polaronic metal to a Fermi Liquid and provide quantum-simulated results of the Fermi-Hubbard model as a benchmark for different theoretical approaches.

Quantum Computing, Communication, and Simulation•Mar 9, 2022•0 citations
Abstract Recent proposals have suggested the use of supervised learning with convolutional neural networks to shed light on some of the less well known phases of the Fermi-Hubbard model through the classification of snapshots from the quantum gas microscopy of ultracold atoms in optical lattices. However, there have been challenges in the interpretability of networks with more than one convolutional filter coupled to the input images. Here, we expand on previous work by considering multiple filters in the first convolutional layer and developing a process for analyzing the physical relevance of patterns obtained in the trained filters. We benchmark our approach at half-filling, where strong antiferromagnetic correlations are known to be present, and we find that upon hole doping, previously unknown patterns arise at temperatures below the tunneling amplitude. These patterns may be a signature of interesting arrangements of fermions in the lattice.
Strongly correlated phases of matter are often described in terms of straightforward electronic patterns. This has so far been the basis for studying the Fermi-Hubbard model realized with ultracold atoms. Here, we show that artificial intelligence (AI) can provide an unbiased alternative to this paradigm for phases with subtle, or even unknown, patterns. Long- and short-range spin correlations spontaneously emerge in filters of a convolutional neural network trained on snapshots of single atomic species. In the less well-understood strange metallic phase of the model, we find that a more complex network trained on snapshots of local moments produces an effective order parameter for the non-Fermi-liquid behavior. Our technique can be employed to characterize correlations unique to other phases with no obvious order parameters or signatures in projective measurements, and has implications for science discovery through AI beyond strongly correlated systems.
Image-like data from quantum systems promises to offer greater insight into the physics of correlated quantum matter. However, the traditional framework of condensed matter physics lacks principled approaches for analyzing such data. Machine learning models are a powerful theoretical tool for analyzing image-like data including many-body snapshots from quantum simulators. Recently, they have successfully distinguished between simulated snapshots that are indistinguishable from one and two point correlation functions. Thus far, the complexity of these models has inhibited new physical insights from such approaches. Here, we develop a set of nonlinearities for use in a neural network architecture that discovers features in the data which are directly interpretable in terms of physical observables. Applied to simulated snapshots produced by two candidate theories approximating the doped Fermi-Hubbard model, we uncover that the key distinguishing features are fourth-order spin-charge correlators. Our approach lends itself well to the construction of simple, versatile, end-to-end interpretable architectures, thus paving the way for new physical insights from machine learning studies of experimental and numerical data.
Although classifying topological quantum phases have attracted great interests, the absence of local order parameter generically makes it challenging to detect a topological phase transition from experimental data. Recent advances in machine learning algorithms enable physicists to analyze experimental data with unprecedented high sensitivities, and identify quantum phases even in the presence of unavoidable noises. Here, we report a successful identification of topological phase transitions using a deep convolutional neural network trained with low signal-to-noise-ratio (SNR) experimental data obtained in a symmetry-protected topological system of spin-orbit-coupled fermions. We apply the trained network to unseen data to map out a whole phase diagram, which predicts the positions of the two topological phase transitions that are consistent with the results obtained by using the conventional method on higher SNR data. By visualizing the filters and post-convolutional results of the convolutional layer, we further find that the CNN uses the same information to make the classification in the system as the conventional analysis, namely spin imbalance, but with an advantage concerning SNR. Our work highlights the potential of machine learning techniques to be used in various quantum systems.
Optical lattice experiments with ultracold atoms allow for the experimental realization of contemporary problems in many-body physics. Yet, devising models that faithfully describe experimental observables is often difficult and problem dependent; there is currently no theoretical method which accounts for all experimental observations. Leveraging the large data volume and presence of strong correlations, machine learning provides a novel avenue for the study of such systems. It has recently been proven successful in analyzing properties of experimental data of ultracold quantum gases. Here we show that generative deep learning succeeds in the challenging task of modeling such an experimental data distribution. Our method is able to produce synthetic experimental snapshots of a doped two-dimensional Fermi-Hubbard model that are indistinguishable from previously reported experimental realizations. We demonstrate how our generative model interprets physical conditions such as temperature at the level of individual configurations. We use our approach to predict snapshots at conditions and scales which are currently experimentally inaccessible, mapping the large-scale behavior of optical lattices at unseen conditions.
Understanding strongly correlated quantum many-body states is one of the most difficult challenges in modern physics. For example, there remain fundamental open questions on the phase diagram of the Hubbard model, which describes strongly correlated electrons in solids. In this work, we realize the Hubbard Hamiltonian and search for specific patterns within the individual images of many realizations of strongly correlated ultracold fermions in an optical lattice. Upon doping a cold-atom antiferromagnet, we find consistency with geometric strings, entities that may explain the relationship between hole motion and spin order, in both pattern-based and conventional observables. Our results demonstrate the potential for pattern recognition to provide key insights into cold-atom quantum many-body systems.
Ultracold neutral atoms in optical lattices have become a powerful platform for the simulation of complex quantum systems. With our quantum gas microscope, we measure snapshots of the quantum many-body wavefunction by fully resolving the position and spin of each atom. I will present the application of this method to two quantum phases: The Haldane spin-1 phase is the prototype of a symmetry-protected topological phase, whose spin order is hidden in conventional two-point correlation functions. On our images, we can evaluate the characteristic string-correlator of this state and directly see the associated edge states of the system. In an antiferromagnetic background, holes get dressed by a surrounding cloud of ferromagnetic correlations. We study these magnetic polarons, which are a building block of the intricate physics of cuprates, by mapping out the spatial structure of the spin environment of individual holes. Upon increased doping, we observe the transition from this unusual polaronic metal to a Fermi Liquid and provide quantum-simulated results of the Fermi-Hubbard model as a benchmark for different theoretical approaches.