Interpreting machine learning of topological quantum phase transitions

Abstract

There has been growing excitement over the possibility of employing artificial neural networks (ANNs) to gain new theoretical insight into the physics of quantum many-body problems. ``Interpretability'' remains a concern: can we understand the basis for the ANN's decision-making criteria in order to inform our theoretical understanding? ``Interpretable'' machine learning in quantum matter has to date been restricted to linear models, such as support vector machines, due to the greater difficulty of interpreting non-linear ANNs. Here we consider topological quantum phase transitions in models of Chern insulator, ${\mathbb{Z}}_2$ topological insulator, and ${\mathbb{Z}}_2$ quantum spin liquid, each using a shallow fully connected feed-forward ANN. The use of quantum loop topography, a ``domain knowledge''-guided approach to feature selection, facilitates the construction of faithful phase diagrams and semi-quantitative interpretation of the criteria in certain cases. To identify the topological phases, the ANNs learn physically meaningful features, such as topological invariants and deconfinement of loops. The interpretability in these cases suggests hope for theoretical progress based on future uses of ANN-based machine learning on quantum many-body problems.

9 pages, 11 figures

References in corpus (5)

• On A Proper Definition of Spin Current
• Machine Learning of Explicit Order Parameters: From the Ising Model to SU(2) Lattice Gauge Theory
• Deep Learning the Quantum Phase Transitions in Random Two-Dimensional Electron Systems
• Unsupervised identification of topological order using predictive models
• Unsupervised learning using topological data augmentation

Cited by in corpus (28)

• Provably efficient machine learning for quantum many-body problems
• Machine Learning for Quantum Matter
• Topological quantum phase transitions retrieved through unsupervised machine learning
• Unsupervised machine learning of topological phase transitions from experimental data
• Discovering Symmetry Invariants and Conserved Quantities by Interpreting Siamese Neural Networks
• Lattice gauge equivariant convolutional neural networks
• Quantitative and Interpretable Order Parameters for Phase Transitions from Persistent Homology
• Machine learning topological invariants of non-Hermitian systems
• Scalable Hamiltonian learning for large-scale out-of-equilibrium quantum dynamics
• Conserved current of nonconserved quantities
• Phase Detection with Neural Networks: Interpreting the Black Box
• Interpretable and unsupervised phase classification
• Topogivity: A Machine-Learned Chemical Rule for Discovering Topological Materials
• Machine learning non-Hermitian topological phases
• Hessian-based toolbox for reliable and interpretable machine learning in physics
• Replacing neural networks by optimal analytical predictors for the detection of phase transitions
• Observing a topological phase transition with deep neural networks from experimental images of ultracold atoms
• Unsupervised Machine Learning of Quenched Gauge Symmetries: A Proof-of-Concept Demonstration
• Unsupervised interpretable learning of topological indices invariant under permutations of atomic bands
• Efficient and quantum-adaptive machine learning with fermion neural networks
• Identifying topology of leaky photonic lattices with machine learning
• Deep learning of topological phase transitions from entanglement aspects for two-dimensional chiral p-wave superconductors
• Quaternion-based machine learning on topological quantum systems
• Fluctuation based interpretable analysis scheme for quantum many-body snapshots
• Learning phases with Quantum Monte Carlo simulation cell
• Discovering quasiorder parameters in the Potts model: A bridge between machine learning and critical phenomena
• Learning quantum symmetries with interactive quantum-classical variational algorithms
• Attention to Quantum Complexity