Related references
Note: Only part of the references are listed.
Article
Astronomy & Astrophysics
Roy T. Forestano et al.
Summary: Recent work has extended deep learning techniques to derive sparse representations of arbitrary Lie algebras, reproducing canonical representations of various Lie groups. The approach is completely general and can be used to find the infinitesimal generators for any Lie group.
Article
Multidisciplinary Sciences
Alexander Roman et al.
Summary: A deep learning methodology is developed to simultaneously discover multiple non-trivial continuous symmetries in a labeled dataset. Symmetry transformations and generators are modeled using fully connected neural networks trained with a specially constructed loss function. The method demonstrates successful identification of symmetries in the data as interpretable non-trivial transformations in a reduced latent space.
Article
Physics, Multidisciplinary
Ziming Liu et al.
Summary: This automated method can find hidden symmetries by quantifying asymmetry and numerically optimizing to minimize violation in all invertible transformations. It has rediscovered hidden symmetries like the Gullstrand-Painlevé metric.
PHYSICAL REVIEW LETTERS
(2022)
Article
Physics, Multidisciplinary
Barry M. Dillon et al.
Summary: This paper introduces the JetCLR method to solve the mapping problem from low-level data to optimized observables through self-supervised contrastive learning. A symmetric data representation is constructed using a permutation-invariant transformer-encoder network for top and QCD jets, and its performance is compared with alternative representations using linear classifier tests, showing good results.
Article
Astronomy & Astrophysics
Sean Craven et al.
Summary: We propose a simple method to identify continuous Lie algebra symmetry in a dataset through regression using an artificial neural network. Our method takes advantage of the scaling of the output variable under infinitesimal symmetry transformations on the input variables and minimizes the possibility of false identification by generating symmetry transformations post-training.
Article
Astronomy & Astrophysics
Krish Desai et al.
Summary: This paper provides a rigorous statistical definition of dataset symmetries and proposes SymmetryGAN, a deep learning method based on generative adversarial networks, for automatically discovering symmetries. It also considers procedures to infer the underlying symmetry group from empirical data.
Article
Physics, Multidisciplinary
Ziming Liu et al.
Summary: AI Poincar6 is a machine learning algorithm that can automatically discover conserved quantities and other properties in unknown dynamical systems. Testing on various Hamiltonian systems shows that it performs well in identifying exact conserved quantities, periodic orbits, phase transitions, and breakdown timescales for approximate conservation laws.
PHYSICAL REVIEW LETTERS
(2021)
Article
Physics, Multidisciplinary
Gabriela Barenboim et al.
Summary: Researchers explore whether Neural Networks can spontaneously discover symmetries as they learn tasks, successfully identifying symmetry information in controlled experiments using Physics templates. This method is also applied to recognize symmetry in artistic paintings.
Article
Physics, Multidisciplinary
Raban Iten et al.
PHYSICAL REVIEW LETTERS
(2020)
Article
Physics, Multidisciplinary
Sebastian J. Wetzel et al.
PHYSICAL REVIEW RESEARCH
(2020)
Review
Physics, Multidisciplinary
Djuna Croon et al.
FRONTIERS IN PHYSICS
(2019)
Article
Physics, Nuclear
Konstantin T. Matchev et al.
JOURNAL OF PHYSICS G-NUCLEAR AND PARTICLE PHYSICS
(2019)
Review
Physics, Nuclear
Jonathan L. Feng
ANNUAL REVIEW OF NUCLEAR AND PARTICLE SCIENCE, VOL 63
(2013)
Article
Optics
A. R. P. Rau
Article
Optics
Robert Zeier et al.
Article
Optics
D. B. Uskov et al.
Article
Physics, Multidisciplinary
T Tilma et al.
JOURNAL OF PHYSICS A-MATHEMATICAL AND GENERAL
(2002)