Machine-learning hidden symmetries

Abstract

We present an automated method for finding hidden symmetries, defined as symmetries that become manifest only in a new coordinate system that must be discovered. Its core idea is to quantify asymmetry as violation of certain partial differential equations, and to numerically minimize such violation over the space of all invertible transformations, parametrized as invertible neural networks. For example, our method rediscovers the famous Gullstrand-Painleve metric that manifests hidden translational symmetry in the Schwarzschild metric of non-rotating black holes, as well as Hamiltonicity, modularity and other simplifying traits not traditionally viewed as symmetries.

Replaced to match accepted PRL version. Improved training, discussion & noise modeling. 14 pages & 4 figs including supplementary material

References in corpus (2)

• E(n) Equivariant Graph Neural Networks
• Machine-Learning Non-Conservative Dynamics for New-Physics Detection

Cited by in corpus (17)

• AI Poincaré 2.0: Machine Learning Conservation Laws from Differential Equations
• Learning Spatiotemporal Chaos Using Next-Generation Reservoir Computing
• Strain topological metamaterials
• Duality of Topological Edge States in a Mechanical Kitaev Chain
• Machine learning understands knotted polymers
• Using machine learning to compress the matter transfer function $T(k)$
• How close Are Integrable and Non-integrable Models: A Parametric Case Study Based on the Salerno Model
• Machine learning unveils the linear matter power spectrum of modified gravity
• Learning phase transitions from regression uncertainty: A new regression-based machine learning approach for automated detection of phases of matter
• Metalearning generalizable dynamics from trajectories
• Discovering Sparse Representations of Lie Groups with Machine Learning
• Analysis of strong coupling constant with machine learning and its application
• Extracting self-similarity from data
• Digital Discovery of a Scientific Concept at the Core of Experimental Quantum Optics
• Quantum Computational Complexity and Symmetry
• Automated detection of symmetry-protected subspaces in quantum simulations
• Deep learning of thermodynamic laws from microscopic dynamics