Discovering Phases, Phase Transitions and Crossovers through Unsupervised Machine Learning: A critical examination

Abstract

We apply unsupervised machine learning techniques, mainly principal component analysis (PCA), to compare and contrast the phase behavior and phase transitions in several classical spin models - the square and triangular-lattice Ising models, the Blume-Capel model, a highly degenerate biquadratic-exchange spin-one Ising (BSI) model, and the 2D XY model, and examine critically what machine learning is teaching us. We find that quantified principal components from PCA not only allow exploration of different phases and symmetry-breaking, but can distinguish phase transition types and locate critical points. We show that the corresponding weight vectors have a clear physical interpretation, which is particularly interesting in the frustrated models such as the triangular antiferromagnet, where they can point to incipient orders. Unlike the other well-studied models, the properties of the BSI model are less well known. Using both PCA and conventional Monte Carlo analysis, we demonstrate that the BSI model shows an absence of phase transition and macroscopic ground-state degeneracy. The failure to capture the `charge' correlations (vorticity) in the BSI model (XY model) from raw spin configurations points to some of the limitations of PCA. Finally, we employ a nonlinear unsupervised machine learning procedure, the `antoencoder method', and demonstrate that it too can be trained to capture phase transitions and critical points.

14 pages, 14 figures

References in corpus (27)

• Deep Learning in Neural Networks: An Overview
• Entanglement Entropy and Quantum Field Theory
• Area laws for the entanglement entropy - a review
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• A quantum gas microscope - detecting single atoms in a Hubbard regime optical lattice
• Entanglement in a simple quantum phase transition
• Machine learning phases of matter
• Single-Atom Resolved Fluorescence Imaging of an Atomic Mott Insulator
• Learning phase transitions by confusion
• Discovering Phase Transitions with Unsupervised Learning
• Single-atom imaging of fermions in a quantum-gas microscope
• Quantum Entanglement in Neural Network States
• Machine learning quantum phases of matter beyond the fermion sign problem
• Introduction to astroML: Machine Learning for Astrophysics
• Learning Thermodynamics with Boltzmann Machines
• Self-Learning Monte Carlo Method
• A Neural Decoder for Topological Codes
• Entropy of entanglement and correlations induced by a quench: Dynamics of a quantum phase transition in the quantum Ising model
• Interplay of quantum and thermal fluctuations in a frustrated magnet
• The Critical Behaviour of the Spin-3/2 Blume-Capel Model in Two Dimensions
• Numerical Linked-Cluster Algorithms. I. Spin systems on square, triangular, and kagome lattices
• PCA meets RG
• Identifying polymer states by machine learning
• First-order phase transition and tricritical scaling behavior of the Blume-Capel model: a Wang-Landau sampling approach
• Phase Diagrams of the Spin-1 Blume-Capel Film With an Alternating Crystal Field
• Monte Carlo Study of an Inhomogeneous Blume-Capel Model
• Projected Regression Methods for Inverting Fredholm Integrals: Formalism and Application to Analytical Continuation

Cited by in corpus (192)

• Machine learning and the physical sciences
• Explainable Machine Learning for Scientific Insights and Discoveries
• Restricted-Boltzmann-Machine Learning for Solving Strongly Correlated Quantum Systems
• Identifying topological order through unsupervised machine learning
• Machine Learning Topological Invariants with Neural Networks
• Constructing neural stationary states for open quantum many-body systems
• Machine learning vortices at the Kosterlitz-Thouless transition
• Unsupervised machine learning and band topology
• Unsupervised machine learning account of magnetic transitions in the Hubbard model
• Machine Learning of Explicit Order Parameters: From the Ising Model to SU(2) Lattice Gauge Theory
• Machine Learning for Quantum Matter
• Quantum Adversarial Machine Learning
• Kernel methods for interpretable machine learning of order parameters
• Unsupervised Learning of Frustrated Classical Spin Models I: Principle Component Analysis
• Neural Quantum States of frustrated magnets: generalization and sign structure
• Identifying Quantum Phase Transitions with Adversarial Neural Networks
• Machine Learning ${\mathbb{Z}}_2$ Quantum Spin Liquids with Quasi-particle Statistics
• Machine learning of phase transitions in the percolation and XY models
• Machine learning of quantum phase transitions
• Topological quantum phase transitions retrieved through unsupervised machine learning
• Topological Quantum Compiling with Reinforcement Learning
• Deep Learning and AdS/CFT
• Machine Learning Many-Body Localization: Search for the Elusive Nonergodic Metal
• Exploring QCD matter in extreme conditions with Machine Learning
• Unsupervised machine learning of topological phase transitions from experimental data
• Parameter diagnostics of phases and phase transition learning by neural networks
• Deep Learning Topological Invariants of Band Insulators
• Scale-invariant Feature Extraction of Neural Network and Renormalization Group Flow
• Self-learning Monte Carlo with Deep Neural Networks
• Probing hidden spin order with interpretable machine learning
• Learning Disordered Topological Phases by Statistical Recovery of Symmetry
• Spectral Reconstruction with Deep Neural Networks
• Machine Learning Bell Nonlocality in Quantum Many-body Systems
• Machine Learning Topological Phases with a Solid-state Quantum Simulator
• Criticality and factorization in the Heisenberg chain with Dzyaloshinskii-Moriya interaction
• Real space mapping of topological invariants using artificial neural networks
• Discovering Symmetry Invariants and Conserved Quantities by Interpreting Siamese Neural Networks
• Measurement-induced criticality as a data-structure transition
• Machine Learning meets Quantum Foundations: A Brief Survey
• Estimating physical properties from liquid crystal textures via machine learning and complexity-entropy methods
• Self-Learning Monte Carlo Method: Continuous-Time Algorithm
• Smallest Neural Network to Learn the Ising Criticality
• Unsupervised machine learning for detection of phase transitions in off-lattice systems I. Foundations
• Unsupervised machine learning of quantum phase transitions using diffusion maps
• Nuclear liquid-gas phase transition with machine learning
• Phase Transition Study meets Machine Learning
• Machine Learning as a universal tool for quantitative investigations of phase transition
• Markovian Quantum Neuroevolution for Machine Learning
• Principal Component Analysis for Fermionic Critical Points
• Automated discovery of characteristic features of phase transitions in many-body localization
• Unsupervised learning universal critical behavior via the intrinsic dimension
• Thermodynamics and Feature Extraction by Machine Learning
• Machine Learning of Frustrated Classical Spin Models. II. Kernel Principal Component Analysis
• The critical temperature of the 2D-Ising model through Deep Learning Autoencoders
• Drawing Phase Diagrams of Random Quantum Systems by Deep Learning the Wave Functions
• dPOLY: Deep Learning of Polymer Phases and Phase Transition
• Towards Novel Insights in Lattice Field Theory with Explainable Machine Learning
• Machine learning discovery of new phases in programmable quantum simulator snapshots
• Quantitative and Interpretable Order Parameters for Phase Transitions from Persistent Homology
• Unsupervised machine learning for detection of phase transitions in off-lattice systems II. Applications
• Unsupervised learning of topological phase transitions using Calinski-Harabaz index
• Divergence of predictive model output as indication of phase transitions
• Machine learning \& artificial intelligence in the quantum domain
• Mapping distinct phase transitions to a neural network
• Deep Learning on the 2-Dimensional Ising Model to Extract the Crossover Region with a Variational Autoencoder
• Interpretable machine learning for inferring the phase boundaries in a nonequilibrium system
• Supervised and unsupervised learning of directed percolation
• Accelerating lattice quantum Monte Carlo simulation using artificial neural networks: an application to the Holstein model
• Neural network setups for a precise detection of the many-body localization transition: finite-size scaling and limitations
• Roadmap on machine learning glassy dynamics
• Topological Persistence Machine of Phase Transitions
• Machine-learning detection of the Berezinskii-Kosterlitz-Thouless transitions in the q-state clock models
• Reducing Autocorrelation Times in Lattice Simulations with Generative Adversarial Networks
• Universal Adversarial Examples and Perturbations for Quantum Classifiers
• Unsupervised and supervised learning of interacting topological phases from single-particle correlation functions
• Extending machine learning classification capabilities with histogram reweighting
• Unsupervised learning using topological data augmentation
• Machine Learning Application to Two-Dimensional Dzyaloshinskii-Moriya Ferromagnets
• Intrinsic dimension of path integrals: data mining quantum criticality and emergent simplicity
• A perspective on machine learning and data science for strongly correlated electron problems
• Self-learning Monte Carlo method with Behler-Parrinello neural networks
• Deep learning of topological phase transitions from entanglement aspects
• Phase transition encoded in neural network
• Quantum kernels to learn the phases of quantum matter
• Few-shot machine learning in the three-dimensional Ising model
• Self-organizing maps as a method for detecting phase transitions and phase identification
• Artificial Neural Network Based Computation for Out-of-Time-Ordered Correlators
• Approximating power of machine-learning ansatz for quantum many-body states
• Machine learning electron correlation in a disordered medium
• Machine learning of phase transitions in nonlinear polariton lattices
• Visualizing Quantum Phases And Identifying Quantum Phase Transitions By Nonlinear Dimensionality Reduction
• A Neural Networks study of the phase transitions of Potts model
• Learning quantum phase transitions through Topological Data Analysis
• First-order transitions and thermodynamic properties in the 2D Blume-Capel model: the transfer-matrix method revisited
• A machine learning approach to the Berezinskii-Kosterlitz-Thouless transition in classical and quantum models
• Identifying structural changes with unsupervised machine learning methods
• Markov Chain Monte Carlo Methods in Quantum Field Theories: A Modern Primer
• Transfer learning of phase transitions in percolation and directed percolation
• Towards meaningful physics from generative models
• A learning by confusion approach to characterize phase transitions
• Machine-learning detection of the Berezinskii-Kosterlitz-Thouless transition and the second-order phase transition in the XXZ models
• Profile approach for recognition of three-dimensional magnetic structures
• Unsupervised machine learning approaches to the $q$-state Potts model
• Extracting many-particle entanglement entropy from observables using supervised machine learning
• Machine learning generated configurations in presence of a conserved quantity: a cautionary tale
• Machine learning for percolation utilizing auxiliary Ising variables
• Unsupervised Learning Eigenstate Phases of Matter
• Visualizing Neural Network Developing Perturbation Theory
• Machine learning phases and criticalities without using real data for training
• Machine learning of XY model on a spherical Fibonacci lattice
• Characterizing the phase diagram of finite-size dipolar Bose-Hubbard systems
• Confinement in non-Abelian lattice gauge theory via persistent homology
• Data-driven discovery of statistically relevant information in quantum simulators
• Supervised and Unsupervised Machine Learning of Structural Phases of Polymers Adsorbed to Nanowires
• Non-parametric learning critical behavior in Ising partition functions: PCA entropy and intrinsic dimension
• Steering-induced phase transition in measurement-only quantum circuits
• Detection of Berezinskii-Kosterlitz-Thouless transition via Generative Adversarial Networks
• Characterizing Different Motility Induced Regimes in Active Matter with Machine Learning and Noise
• Complexity of spin configurations dynamics due to unitary evolution and periodic projective measurements
• Machine learning for structure-property relationships: Scalability and limitations
• Morphology of three-body quantum states from machine learning
• Detecting Depinning and Nonequilibrium Transitions with Unsupervised Machine Learning
• Machine learning phase transitions of the three-dimensional Ising universality class
• Study of phase transition of Potts model with Domain Adversarial Neural Network
• Detection of the onset of yielding and creep failure from digital image correlation
• Probing the transition from dislocation jamming to pinning by machine learning
• Examples of renormalization group transformations for image sets
• Exponentially improved efficient machine learning for quantum many-body states with provable guarantees
• An analysis of localization transitions using non-parametric unsupervised learning
• Unsupervised Machine Learning of Quenched Gauge Symmetries: A Proof-of-Concept Demonstration
• Network science Ising states of matter
• Adversarial Machine Learning Phases of Matter
• Probing Criticality in Quantum Spin Chains with Neural Networks
• Detecting composite orders in layered models via machine learning
• Unsupervised Learning of Rydberg Atom Array Phase Diagram with Siamese Neural Networks
• Principal component analysis of absorbing state phase transitions
• Direct detection of plasticity onset through total-strain profile evolution
• A simple framework for contrastive learning phases of matter
• Emergence of a finite-size-scaling function in the supervised learning of the Ising phase transition
• Deep Learning of Phase Transitions for Quantum Spin Chains from Correlation Aspects
• Machine Learning approach to muon spectroscopy analysis
• Machine learning phases of an Abelian gauge theory
• Learning by Confusion: The Phase Diagram of the Holstein Model
• Comprehensive studies on the universality of BKT transitions -- Machine-learning study, Monte Carlo simulation, and Level-spectroscopy method
• Supervised and unsupervised learning of the many-body critical phase, phase transitions, and critical exponents in disordered quantum systems
• Deep learning of phase transitions with minimal examples
• Rapid detection of phase transitions from Monte Carlo samples before equilibrium
• Thermodynamics of the Ising model encoded in restricted Boltzmann machines
• Machine learning classification of two-dimensional vortex configurations
• Extracting Off-Diagonal Order from Diagonal Basis Measurements
• Finding Quantum Critical Points with Neural-Network Quantum States
• Human-machine collaboration: ordering mechanism of rank-2 spin liquid on breathing pyrochlore lattice
• Learning What a Machine Learns in a Many-Body Localization Transition
• Detecting the phase transition in a strongly-interacting Fermi gas by unsupervised machine learning
• Application of the Variational Autoencoder to Detect the Critical Points of the Anisotropic Ising Model
• Predicting nucleation near the spinodal in the Ising model using machine learning
• Deep learning of topological phase transitions from entanglement aspects for two-dimensional chiral p-wave superconductors
• Principal Component Analysis of Diffuse Magnetic Scattering: a Theoretical Study
• Learning phases with Quantum Monte Carlo simulation cell
• Phase classification in the long-range Harper model using machine learning
• Unsupervised Machine Learning Phase Classification for Falicov-Kimball Model
• Machine learning, quantum chaos, and pseudorandom evolution
• Machine Learning Domain Adaptation in Spin Models with Continuous Phase Transitions
• Correlations in the shear flow of athermal amorphous solids: A principal component analysis
• Supervised learning of an interacting 2D hard-core boson model of a weak topological insulator using correlation functions
• Deep Learning Super-Diffusion in Multiplex Networks
• Many-body mobility edges in 1D and 2D revealed by convolutional neural networks
• Exploring percolation phase transition in the three-dimensional Ising model with machine learning
• Confusion-driven machine learning of structural phases of a flexible, magnetic Stockmayer polymer
• Weighted Quantum Channel Compiling through Proximal Policy Optimization
• Learning phase transitions by siamese neural network
• Diagnosing quantum transport from wave function snapshots
• Detecting the Largest Correlations using the Correlation Density Matrix: a Quantum Monte Carlo Approach
• Physical meaning of principal component analysis for classical lattice systems with translational invariance
• Machine-Learning Detection of the Berezinskii-Kosterlitz-Thouless Transitions
• Relevant Analytic Spontaneous Magnetization Relation for the Face-Centered-Cubic Ising Lattice
• Deep learning of topological phase transitions from entanglement aspects: An unsupervised way
• Employing machine learning for theory validation and identification of experimental conditions in laser-plasma physics
• Machine learning that predicts well may not learn the correct physical descriptions of glassy systems
• Characterizing spin ordering via maximal row correlation in classical spin models
• Discovering quasiorder parameters in the Potts model: A bridge between machine learning and critical phenomena
• Neural Scaling Laws for Deep Regression
• Unsupervised Learning of Symmetry Protected Topological Phase Transitions
• Global exploration of phase behavior in frustrated Ising models using unsupervised learning techniques
• Efficiency of neural-network state representations of one-dimensional quantum spin systems
• Optimized Observable Readout from Single-shot Images of Ultracold Atoms via Machine Learning
• Orthogonal projections of hypercubes
• Snake net and balloon force with a neural network for detecting multiple phases
• The statistical mechanics and machine learning of the $\alpha$-Rényi ensemble
• Uncovering Magnetic Phases with Synthetic Data and Physics-Informed Training
• Machine learning analysis of dimensional reduction conjecture for nonequilibrium Berezinskii-Kosterlitz-Thouless transition in three dimensions
• Unsupervised classification of disordered patterns in an oppositely charged colloidal system