Characterizing out-of-distribution generalization of neural networks: application to the disordered Su-Schrieffer-Heeger model
arXiv:2406.10012 · doi:10.1088/2632-2153/ad9079
Abstract
Machine learning (ML) is a promising tool for the detection of phases of matter. However, ML models are also known for their black-box construction, which hinders understanding of what they learn from the data and makes their application to novel data risky. Moreover, the central challenge of ML is to ensure its good generalization abilities, i.e., good performance on data outside the training set. Here, we show how the informed use of an interpretability method called class activation mapping (CAM), and the analysis of the latent representation of the data with the principal component analysis (PCA) can increase trust in predictions of a neural network (NN) trained to classify quantum phases. In particular, we show that we can ensure better out-of-distribution generalization in the complex classification problem by choosing such an NN that, in the simplified version of the problem, learns a known characteristic of the phase. We show this on an example of the topological Su-Schrieffer-Heeger (SSH) model with and without disorder, which turned out to be surprisingly challenging for NNs trained in a supervised way. This work is an example of how the systematic use of interpretability methods can improve the performance of NNs in scientific problems.
arXiv comment23 pages, 13 figures, example code is available at https://github.com/kcybinski/Interpreting_NNs_for_topological_phases_of_matter
References in corpus (65)
- Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
- Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks
- Quantum machine learning in feature Hilbert spaces
- Machine learning phases of matter
- Learning phase transitions by confusion
- Discovering Phase Transitions with Unsupervised Learning
- Unsupervised learning of phase transitions: from principal component analysis to variational autoencoders
- Generalization in quantum machine learning from few training data
- Machine learning quantum phases of matter beyond the fermion sign problem
- Observation of the topological Anderson insulator in disordered atomic wires
- Topological Criticality in the Chiral-Symmetric AIII Class at Strong Disorder
- Identifying topological order through unsupervised machine learning
- Machine Learning Topological Invariants with Neural Networks
- Identifying Quantum Phase Transitions using Artificial Neural Networks on Experimental Data
- Unsupervised machine learning and band topology
- Artificial Intelligence and Machine Learning for Quantum Technologies
- 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
- Exact diagonalization: the Bose-Hubbard model as an example
- Ab-initio quantum chemistry with neural-network wavefunctions
- Identifying Quantum Phase Transitions with Adversarial Neural Networks
- A Separability-Entanglement Classifier via Machine Learning
- Unsupervised phase discovery with deep anomaly detection
- From Clustering to Cluster Explanations via Neural Networks
- Discriminative Cooperative Networks for Detecting Phase Transitions
- Machine learning of quantum phase transitions
- Unsupervised Manifold Clustering of Topological Phononics
- Topological quantum phase transitions retrieved through unsupervised machine learning
- Unsupervised machine learning of topological phase transitions from experimental data
- Probing hidden spin order with interpretable machine learning
- Extrapolating quantum observables with machine learning: Inferring multiple phase transitions from properties of a single phase
- Unsupervised identification of topological order using predictive models
- Discovering Symmetry Invariants and Conserved Quantities by Interpreting Siamese Neural Networks
- Learning multiple order parameters with interpretable machines
- Topological phases of a dimerized Fermi-Hubbard model for semiconductor nano-lattices
- Correlator Convolutional Neural Networks: An Interpretable Architecture for Image-like Quantum Matter Data
- Applications of neural networks to the studies of phase transitions of two-dimensional Potts models
- Automated discovery of characteristic features of phase transitions in many-body localization
- Towards Novel Insights in Lattice Field Theory with Explainable Machine Learning
- Quantitative and Interpretable Order Parameters for Phase Transitions from Persistent Homology
- Machine learning discovery of new phases in programmable quantum simulator snapshots
- Mapping distinct phase transitions to a neural network
- Divergence of predictive model output as indication of phase transitions
- Neural-network quantum states for many-body physics
- Visualizing Strange Metallic Correlations in the 2D Fermi-Hubbard Model with AI
- Neural network setups for a precise detection of the many-body localization transition: finite-size scaling and limitations
- Topological properties of the long-range Kitaev chain with Aubry-André-Harper modulation
- Model-Independent Learning of Quantum Phases of Matter with Quantum Convolutional Neural Networks
- Topological data analysis and machine learning
- Unsupervised and supervised learning of interacting topological phases from single-particle correlation functions
- Revealing the Phase Diagram of Kitaev Materials by Machine Learning: Cooperation and Competition between Spin Liquids
- Phase Detection with Neural Networks: Interpreting the Black Box
- Deep learning of topological phase transitions from entanglement aspects
- Interpretable and unsupervised phase classification
- Visualizing Quantum Phases And Identifying Quantum Phase Transitions By Nonlinear Dimensionality Reduction
- Identification of hidden order and emergent constraints in frustrated magnets using tensorial kernel methods
- Detecting ergodic bubbles at the crossover to many-body localization using neural networks
- A supervised learning algorithm for interacting topological insulators based on local curvature
- Unsupervised learning of topological phase diagram using topological data analysis
- A learning by confusion approach to characterize phase transitions
- Unsupervised mapping of phase diagrams of 2D systems from infinite projected entangled-pair states via deep anomaly detection
- Entanglement Classification via Witness Operators generated by Support Vector Machine
- Unsupervised Learning of Rydberg Atom Array Phase Diagram with Siamese Neural Networks
- Unsupervised interpretable learning of topological indices invariant under permutations of atomic bands
- Classifying topological neural network quantum states via diffusion maps